1
00:00:05,280 --> 00:00:10,600
[CLAIRE] Welcome to Talking Postgres. This is a
monthly podcast for developers who love this database.

2
00:00:10,820 --> 00:00:12,380
I'm your host, Claire Giordano,

3
00:00:12,840 --> 00:00:18,620
and in this pod, we explore the human side
of Postgres databases and open source, which

4
00:00:18,680 --> 00:00:21,620
means why do people who work
with Postgres do what they do,

5
00:00:22,100 --> 00:00:23,200
and how did they get there?

6
00:00:23,840 --> 00:00:28,280
I want to say thank you to the team at Microsoft
for sponsoring today's community conversation.

7
00:00:29,300 --> 00:00:37,020
And, drumroll, today's guest is Karen Jex. Karen
is a senior software engineer at Snowflake, which

8
00:00:37,580 --> 00:00:42,200
she formerly worked at Crunchy Data,
which was acquired by Snowflake in 2025,

9
00:00:42,840 --> 00:00:44,960
and she's officially recognized as a

10
00:00:45,020 --> 00:00:52,480
significant Postgres contributor. Karen serves
on the organizing team for PGConf.EU 2026,

11
00:00:52,960 --> 00:00:56,440
and she leads the diversity committee
for PostgreSQL Europe as well.

12
00:00:57,100 --> 00:01:03,840
And she has worked in databases since
1998 and in Postgres since 2019.

13
00:01:04,019 --> 00:01:06,140
She's a frequent Postgres conference speaker

14
00:01:06,740 --> 00:01:09,900
and talk selection team member
as well, a mountain biker.

15
00:01:10,460 --> 00:01:14,140
I think mountain biking might have a
lot to do with why she lives in the

16
00:01:14,180 --> 00:01:17,960
country she lives in. And with
that, I want to say welcome.

17
00:01:17,860 --> 00:01:22,580
[KAREN] Thank you, Claire. It's a real
pleasure to be here. I'm really happy.

18
00:01:24,420 --> 00:01:28,980
[CLAIRE] Today's topic is from
Oracle DBA to Postgres contributor.

19
00:01:29,920 --> 00:01:35,620
And I actually, as we were discussing
title ideas and episode focus,

20
00:01:36,180 --> 00:01:39,840
I didn't realize that you were
an Oracle DBA once upon a time.

21
00:01:41,200 --> 00:01:42,760
[KAREN] For about 20 years.

22
00:01:43,760 --> 00:01:47,260
[CLAIRE] Wow. OK, so how did
you, let's start with that.

23
00:01:47,340 --> 00:01:50,380
I guess we're not gonna get to Postgres
for a while, but hopefully we'll get there

24
00:01:50,440 --> 00:01:54,980
soon. How did you get started in
databases? What was that path?

25
00:01:55,787 --> 00:02:00,580
[KAREN] I like quite a few things
in my career completely by accident.

26
00:02:01,500 --> 00:02:04,360
So I did a maths degree.

27
00:02:05,880 --> 00:02:11,980
And I actually was going to start doing a
PGCE, which is teacher training in England.

28
00:02:12,960 --> 00:02:15,440
But my parents were both teachers,

29
00:02:16,060 --> 00:02:20,480
and we had various other family
members who were teachers, and I saw

30
00:02:22,360 --> 00:02:24,500
what it did to them, basically,

31
00:02:25,340 --> 00:02:29,520
and decided that I really, really didn't
think I could do that. So I had to

32
00:02:30,040 --> 00:02:31,060
choose something else to do.

33
00:02:31,720 --> 00:02:36,500
And at the kind of last
minute, I thought, actually,

34
00:02:37,500 --> 00:02:40,280
I'm going to do a software development MSc.

35
00:02:41,640 --> 00:02:48,460
So I signed up for a software
development MSc, and part of that was

36
00:02:49,260 --> 00:02:55,360
database theory, so relational databases, data
modeling, data flow diagrams, that kind of thing.

37
00:02:56,700 --> 00:02:58,420
And that was my favorite part

38
00:02:59,600 --> 00:03:04,660
of that course, but I still
didn't have any idea about

39
00:03:06,460 --> 00:03:08,900
what kind of jobs existed
in the world of databases?

40
00:03:10,360 --> 00:03:13,280
So then when I started applying for graduate jobs,

41
00:03:14,040 --> 00:03:17,800
out of that, I applied for the graduate

42
00:03:19,640 --> 00:03:23,300
IT scheme at Barnardo's, the
children's charity in England.

43
00:03:24,140 --> 00:03:25,740
And they did aptitude testing.

44
00:03:26,480 --> 00:03:30,740
So as well as interviews, we were
sat down in a room to do a whole

45
00:03:30,820 --> 00:03:32,360
series of aptitude tests,

46
00:03:33,080 --> 00:03:36,980
and they came back from those saying, Well,
we think you'd be a really good database

47
00:03:37,040 --> 00:03:39,020
administrator. We'd like to offer you a job

48
00:03:39,780 --> 00:03:41,140
as a database administrator.

49
00:03:42,600 --> 00:03:45,940
And then I had to go back
to my dad, who was in IT,

50
00:03:46,680 --> 00:03:52,120
to say, what is a database administrator?
Because Google didn't exist then. I couldn't just

51
00:03:53,000 --> 00:03:54,320
look up online what that meant.

52
00:03:55,120 --> 00:03:57,040
So he explained what it was, and

53
00:03:58,160 --> 00:04:00,500
I thought, yeah, okay, that sounds
like the kind of thing I'd enjoy.

54
00:04:01,620 --> 00:04:04,920
And it must have been because they must have

55
00:04:05,400 --> 00:04:09,900
got it right because I carried on doing
pretty much exactly the same thing for

56
00:04:10,360 --> 00:04:11,220
the next 20 years.

57
00:04:12,700 --> 00:04:17,240
[CLAIRE] And was this all with Oracle in
those early years, or in those 20 years, I

58
00:04:17,279 --> 00:04:17,660
should say?

59
00:04:18,960 --> 00:04:25,260
[KAREN] Yes, so through my MSc,
we didn't even have access to any

60
00:04:26,660 --> 00:04:29,240
kind of real databases, literally I used

61
00:04:30,140 --> 00:04:32,940
Access to create my original,

62
00:04:33,780 --> 00:04:40,400
like, first databases when I was studying, I didn't
know anything about any other databases then.

63
00:04:40,860 --> 00:04:45,020
So when I started my first job as a
database administrator, they sent me off to

64
00:04:45,060 --> 00:04:47,080
do all of the Oracle training courses.

65
00:04:48,280 --> 00:04:55,440
So I got my Oracle DBA certification, and
I kept renewing that through the years,

66
00:04:56,780 --> 00:04:59,260
and we only had Oracle databases there.

67
00:05:04,160 --> 00:05:09,140
I think my next two jobs as a database
administrator, again, it was just Oracle.

68
00:05:11,580 --> 00:05:15,320
There was a bit of SQL Server and
a bit of Sybase kind of got mixed

69
00:05:15,400 --> 00:05:16,760
in somewhere along the line.

70
00:05:17,960 --> 00:05:20,280
But then, kind of back to Oracle.

71
00:05:22,280 --> 00:05:24,280
Yeah, till about 2019.

72
00:05:25,200 --> 00:05:32,180
[CLAIRE] Okay, so what dramatic event
in 2019 changed your focus on databases?

73
00:05:32,860 --> 00:05:35,000
Tell me there's a story there. Is there a story?

74
00:05:36,860 --> 00:05:40,880
[KAREN] It's not necessarily a super
exciting story, but it's another of those

75
00:05:41,460 --> 00:05:48,860
complete accident type things. So I,
for seven years, I was working remotely

76
00:05:49,680 --> 00:05:52,500
for the company I used to
work for in London, so I was

77
00:05:53,220 --> 00:05:57,500
contracting as a DBA for them. That was
supposed to be a five-month contract.

78
00:05:58,060 --> 00:06:00,160
It was one of those that
ended up being seven years.

79
00:06:00,580 --> 00:06:04,980
And then finally they couldn't justify keeping
me on as a contractor in France when they

80
00:06:05,000 --> 00:06:08,340
were doing their things in
other places. So I had to

81
00:06:09,344 --> 00:06:12,464
decide what I was going to do next, and I found

82
00:06:14,804 --> 00:06:20,284
a contract in France as a so-called
database expert on a team.

83
00:06:21,604 --> 00:06:25,204
And it was actually on the project
that was rolling out the smart meters,

84
00:06:25,864 --> 00:06:27,584
smart electricity meters across France.

85
00:06:28,544 --> 00:06:29,724
And that was all in Oracle.

86
00:06:30,604 --> 00:06:35,164
So I got that contract and I worked
with the team on this project to roll

87
00:06:35,224 --> 00:06:39,264
out the smart meters across France, and then

88
00:06:40,384 --> 00:06:44,384
a year or two into that, they decided
they were going to do a new project,

89
00:06:45,024 --> 00:06:46,484
and it was going to be in Postgres.

90
00:06:46,724 --> 00:06:50,864
Now I still, to this day, have no
idea who decided Postgres because

91
00:06:51,484 --> 00:06:54,064
nobody on the project knew
anything about Postgres.

92
00:06:55,084 --> 00:06:58,244
We didn't know how to download it.
We didn't know how to install it.

93
00:06:58,284 --> 00:07:00,004
We didn't know how to back it up.

94
00:07:00,044 --> 00:07:02,884
We didn't know how to put
high availability in place.

95
00:07:04,204 --> 00:07:06,844
But they just said, Oh, well, you're the
database expert. You can figure it out.

96
00:07:05,710 --> 00:07:10,770
[CLAIRE] So, did it feel like
it was a top-down decision?

97
00:07:10,950 --> 00:07:14,970
Like somebody decided either they wanted to
go with open source or they wanted to reduce

98
00:07:15,050 --> 00:07:18,690
costs, or you never know why.
You still to this day don't know.

99
00:07:19,544 --> 00:07:21,924
[KAREN] I strongly suspect
they wanted to reduce costs.

100
00:07:22,844 --> 00:07:26,264
But the reasons were never
communicated. It was very much just a—

101
00:07:26,264 --> 00:07:32,504
This project's going to use Postgres,
and deal with it, yeah, yeah.

102
00:07:29,410 --> 00:07:30,530
[CLAIRE] Deal with it. Go learn.

103
00:07:32,864 --> 00:07:37,704
[KAREN] So I had to go and figure out
what Postgres was and how to use it.

104
00:07:37,964 --> 00:07:42,884
And then as soon as I figured
out how to use it, I had to then

105
00:07:42,964 --> 00:07:49,264
train the developers on Postgres, pretending that
I was, you know, this Postgres expert. So yeah,

106
00:07:49,884 --> 00:07:51,044
it was good. It was good fun.

107
00:07:52,630 --> 00:07:57,490
[CLAIRE] Now, you mentioned something about living in
France and working remotely for that seven-year period

108
00:07:57,530 --> 00:08:02,730
back in those earlier Oracle days of
your life. Why were you living in France?

109
00:08:02,830 --> 00:08:06,070
I think this might be where the mountain
biker part of the story comes in.

110
00:08:07,184 --> 00:08:09,784
[KAREN] It is exactly where the
mountain biking part comes in.

111
00:08:10,364 --> 00:08:15,244
My husband and I have both
always loved cycling and,

112
00:08:16,324 --> 00:08:20,004
yeah, all of our holidays and
some of our weekends were spent

113
00:08:20,524 --> 00:08:24,384
finding places to do mountain biking.
We lived in the southeast of England,

114
00:08:24,904 --> 00:08:30,564
which is very much not known for its
mountains, not even hills, really.

115
00:08:32,584 --> 00:08:34,024
And then we

116
00:08:34,824 --> 00:08:39,344
came on a holiday to a town called
Morzine, which is near where we now live,

117
00:08:39,984 --> 00:08:41,104
for a mountain bike holiday.

118
00:08:41,824 --> 00:08:45,244
And a couple of weeks after we got
home and were driving on one of the

119
00:08:46,084 --> 00:08:50,144
UK motorways, we just kind of looked at
each other and said, like, wouldn't it be

120
00:08:50,204 --> 00:08:50,864
great if we could

121
00:08:52,044 --> 00:08:54,264
just go back and live in Morzine, and

122
00:08:55,164 --> 00:09:00,184
I think by a year, just over a
year later, we'd sold up in the UK.

123
00:09:00,284 --> 00:09:00,464
We'd

124
00:09:01,084 --> 00:09:04,684
quit our jobs and we bought the
house in France and moved here.

125
00:09:06,770 --> 00:09:07,190
[CLAIRE] Very cool.

126
00:09:08,530 --> 00:09:10,250
I can't imagine.

127
00:09:10,710 --> 00:09:14,090
I mean, you have some real mountains
in the town that you live in.

128
00:09:14,750 --> 00:09:19,150
So when you go off and do or did
mountain bike rides, I mean, you're not

129
00:09:19,210 --> 00:09:21,110
just talking about small hills, right?

130
00:09:22,470 --> 00:09:23,030
On these.

131
00:09:22,764 --> 00:09:26,264
[KAREN] No, we're halfway
up a mountain, basically.

132
00:09:26,364 --> 00:09:32,584
We live at 820 meters altitude, and then
the mountain kind of, you can see it

133
00:09:32,664 --> 00:09:33,104
out of our

134
00:09:34,244 --> 00:09:36,824
window, out of my office
window, you see the mountain.

135
00:09:38,304 --> 00:09:41,304
Yeah, so our village has its
own little ski resort, so we can

136
00:09:41,864 --> 00:09:45,404
drive for five minutes up the road and
go skiing in the winter, which is really

137
00:09:45,464 --> 00:09:45,744
cool.

138
00:09:46,470 --> 00:09:48,150
[CLAIRE] That is really cool. I like it.

139
00:09:49,430 --> 00:09:53,890
And thank goodness that people can,
I don't know, work remote these days

140
00:09:55,250 --> 00:09:58,570
in order to do this kind of thing,
live in a beautiful place like that.

141
00:09:55,704 --> 00:09:56,004
[KAREN] Yeah, well.

142
00:09:59,664 --> 00:10:03,684
When we first moved here, there was no broadband.

143
00:10:04,544 --> 00:10:08,624
So I did have to give up my job for
a while. I couldn't work remotely.

144
00:10:10,644 --> 00:10:16,664
So I did several other things whilst we
waited for broadband to be installed.

145
00:10:18,344 --> 00:10:23,144
Yeah, so I'm not sure, looking back,
what made us think it was a good idea.

146
00:10:24,264 --> 00:10:27,984
Because we had no idea how we
were going to support ourselves.

147
00:10:29,324 --> 00:10:31,264
I had a career that I really enjoyed,

148
00:10:32,824 --> 00:10:36,724
that I knew I was going to have
to let go of for a little while.

149
00:10:38,884 --> 00:10:39,984
Yeah, it worked.

150
00:10:42,830 --> 00:10:47,890
[CLAIRE] So you're working for this smart
meter company, rolling out smart meters.

151
00:10:48,730 --> 00:10:52,190
Leveraging all of your Oracle
expertise, that sounds like you had

152
00:10:52,870 --> 00:10:56,630
more than 15 years under your
belt at this point. They decide,

153
00:10:57,190 --> 00:11:01,230
for some unknown reason, to switch to
Postgres. You had to figure it out.

154
00:11:01,450 --> 00:11:06,290
You had to learn. And then what was your role?

155
00:11:02,364 --> 00:11:02,764
[KAREN] Yes.

156
00:11:06,470 --> 00:11:09,670
[CLAIRE] You were a DBA at that
point still, then with Postgres?

157
00:11:10,484 --> 00:11:15,924
[KAREN] So they called me the
database expert, which was kind of...

158
00:11:17,024 --> 00:11:23,824
Yeah, it was really a DBA role, but
supporting the project with whatever.

159
00:11:24,944 --> 00:11:29,624
Database knowledge, etc. they needed. Oh,
that new project as well, they decided

160
00:11:30,864 --> 00:11:35,504
they were going to use Ansible to roll
everything out that equally I had no idea

161
00:11:35,584 --> 00:11:40,164
about. So I had to not only learn
how to install Postgres, but also how

162
00:11:42,284 --> 00:11:44,784
to make that work on Ansible
so that we could automate

163
00:11:47,284 --> 00:11:49,244
the setup and everything. So yeah, that was fun.

164
00:11:50,330 --> 00:11:56,430
[CLAIRE] So, do you remember any of those first
impressions or first experiences with Postgres?

165
00:11:56,810 --> 00:12:01,470
What did you think of the documentation? Were
you happy with the getting started materials?

166
00:12:02,270 --> 00:12:04,390
Did you, how did you discover

167
00:12:05,670 --> 00:12:10,690
the open source community and the conferences,
or, I don't know, step me through that getting

168
00:12:10,750 --> 00:12:12,790
indoctrinated into the Postgres world.

169
00:12:14,580 --> 00:12:21,040
[KAREN] So I started, yeah, I started
with the Postgres docs, and that was—

170
00:12:22,160 --> 00:12:27,340
That was a pretty good experience for
getting up and running with Postgres.

171
00:12:28,020 --> 00:12:30,380
The bit that I struggled
with, and I think some people

172
00:12:30,840 --> 00:12:34,260
still struggle now, although it is getting
better and people are trying really hard

173
00:12:34,760 --> 00:12:35,920
to make it easier, was then

174
00:12:36,620 --> 00:12:40,960
how to put things together, because I came
from an Oracle world where everything's kind of

175
00:12:41,620 --> 00:12:45,020
bundled together for you. You
don't just have an Oracle database.

176
00:12:45,120 --> 00:12:49,300
You've got your backup and recovery, you've got
your replication, you've got your high availability,

177
00:12:49,360 --> 00:12:50,260
you've got everything

178
00:12:51,320 --> 00:12:54,420
built in for you, and I know that comes
with a price, but fortunately I was

179
00:12:54,440 --> 00:12:57,400
never the one signing the checks. I
just got to play with the tech. So

180
00:12:58,200 --> 00:13:01,240
from my point of view, that was
great. So then with Postgres,

181
00:13:02,640 --> 00:13:04,180
I had to figure out that

182
00:13:04,920 --> 00:13:05,600
we don't have

183
00:13:06,520 --> 00:13:09,060
everything there in one
place. You have to build it

184
00:13:09,940 --> 00:13:13,200
for yourself, you have to put all the
different pieces together. So that was a real

185
00:13:14,160 --> 00:13:15,080
culture shock for me.

186
00:13:17,760 --> 00:13:19,400
And it took me a while.

187
00:13:20,160 --> 00:13:24,000
You know, I managed to get my
first databases up and running and

188
00:13:25,200 --> 00:13:27,440
being backed up and practice recovery.

189
00:13:28,880 --> 00:13:33,160
And then it took me a good while longer
after that, I would say months, to

190
00:13:33,480 --> 00:13:38,080
fully get my head around
things like high availability.

191
00:13:41,500 --> 00:13:45,720
Connection pooling, all the
other things that go around it.

192
00:13:45,600 --> 00:13:51,940
[CLAIRE] I actually was in New York City for
Postgres Summit US a couple weeks ago, and

193
00:13:52,020 --> 00:13:54,960
I gave a beginner's guide talk about replication.

194
00:13:55,620 --> 00:13:58,460
And it was, the way I described it is it was

195
00:13:59,240 --> 00:14:03,160
the talk I wish I had seen when I
first got involved with Postgres,

196
00:14:03,620 --> 00:14:08,500
because I felt like all the different types
of replication were an alphabet soup, right?

197
00:14:08,760 --> 00:14:13,240
I mean, there was shipping and streaming
and synchronous and asynchronous and...

198
00:14:13,840 --> 00:14:17,400
I, you know, logical and physical.

199
00:14:16,400 --> 00:14:17,880
[KAREN] Yeah, logical and physical.

200
00:14:18,460 --> 00:14:23,920
[CLAIRE] And so anyway, I just kind of set
people through it to explain these things and

201
00:14:24,260 --> 00:14:29,899
tease them apart and realize there's really
just two main forms of replication. But yeah, I

202
00:14:31,720 --> 00:14:36,260
think people have been improving that beginner's
guide level information over the years.

203
00:14:36,420 --> 00:14:37,100
I think you would.

204
00:14:36,940 --> 00:14:38,100
[KAREN] Yes, I think so.

205
00:14:39,500 --> 00:14:42,260
[CLAIRE] But I don't know if we're
100% there yet. We're getting there.

206
00:14:43,780 --> 00:14:47,420
[KAREN] I think we're getting there and we
listen and we're trying to make it better.

207
00:14:48,360 --> 00:14:48,960
[CLAIRE] Absolutely.

208
00:14:49,820 --> 00:14:50,640
Okay, so

209
00:14:51,520 --> 00:14:54,620
this was 2019

210
00:14:55,760 --> 00:14:57,720
that you first started getting involved

211
00:14:58,820 --> 00:15:03,020
in Postgres. And I know there are people
out there who work with Oracle today

212
00:15:03,500 --> 00:15:08,300
who are probably listening to this episode,
and they have not yet started to build up

213
00:15:08,360 --> 00:15:09,460
their Postgres skills.

214
00:15:09,980 --> 00:15:13,660
And so as you reflect back on your
experience, and I didn't prepare you for this

215
00:15:13,720 --> 00:15:18,220
question, so maybe you might decide, no, Claire,

216
00:15:19,020 --> 00:15:22,440
I'll send you some links after we hang
up, after the episode is recorded.

217
00:15:22,800 --> 00:15:27,320
But are there any particular resources
that you would point an Oracle person to

218
00:15:27,940 --> 00:15:30,020
say, you should do this or you should read that?

219
00:15:31,420 --> 00:15:32,200
Any favorites?

220
00:15:33,020 --> 00:15:34,560
[KAREN] I did find, and I can't

221
00:15:35,020 --> 00:15:38,760
for the life of me remember what
the link was, but I did find a guide

222
00:15:38,800 --> 00:15:43,900
that someone had written, which
was Postgres for Oracle DBAs.

223
00:15:45,840 --> 00:15:50,000
And yeah, I can't remember where it was,
but it was quite handy because it just

224
00:15:50,060 --> 00:15:50,840
kind of gave a—

225
00:15:53,540 --> 00:15:55,340
a kind of, oh, you know, if you know this

226
00:15:56,420 --> 00:15:58,340
mechanism in Oracle, then

227
00:15:59,340 --> 00:16:02,000
that's what it is in Postgres, and
there were some things like that.

228
00:16:03,180 --> 00:16:07,340
Oh, there you go, Postgres.
It's probably that one, yes.

229
00:16:07,240 --> 00:16:10,980
[CLAIRE] Well, we'll put it in the show
notes, whatever the right one is. It's,

230
00:16:11,480 --> 00:16:15,700
we'll, I'll talk to Karen after the show.
We'll find the link and we'll include it.

231
00:16:16,960 --> 00:16:19,700
Because, yeah, I think it helps sometimes when

232
00:16:20,380 --> 00:16:25,240
people find a resource that's geared toward whatever
the mental model is they're already familiar

233
00:16:25,280 --> 00:16:28,700
with, and then connects the dots from there.

234
00:16:25,480 --> 00:16:26,620
[KAREN] Yes, exactly.

235
00:16:31,340 --> 00:16:36,659
[CLAIRE] Okay, so when I met you were
not working for that smart meter company.

236
00:16:36,760 --> 00:16:42,760
You were working for Crunchy Data as a
solution architect, and you were 100% on

237
00:16:43,500 --> 00:16:45,420
Postgres at that point in time, right?

238
00:16:45,800 --> 00:16:46,160
[KAREN] Yes.

239
00:16:47,480 --> 00:16:49,300
So it didn't take long

240
00:16:49,760 --> 00:16:56,320
after I discovered Postgres in 2019 for me
to move completely to Postgres full-time.

241
00:16:56,440 --> 00:16:57,540
In fact, from

242
00:16:59,080 --> 00:17:02,120
January 2020, I was working
with Postgres full time.

243
00:17:05,600 --> 00:17:06,800
So it was pretty fast.

244
00:17:07,840 --> 00:17:10,180
[CLAIRE] One of the things I'm curious about is

245
00:17:11,740 --> 00:17:13,400
solution architect versus DBA.

246
00:17:13,880 --> 00:17:17,040
Those are two different, are they two
different job titles for the same thing?

247
00:17:17,820 --> 00:17:21,500
Or are they two completely different
roles? Because you've worn both hats.

248
00:17:23,000 --> 00:17:26,320
[KAREN] Yeah, so I went first to being a

249
00:17:27,020 --> 00:17:30,980
Postgres consultant, so I was
kind of helping people to—

250
00:17:32,420 --> 00:17:38,220
I always said that that was going from
being a DBA to helping other DBAs.

251
00:17:39,880 --> 00:17:44,660
And then I moved into the solution architect
role, and I loved that role at Crunchy

252
00:17:44,720 --> 00:17:48,060
Data because it was a hybrid role.

253
00:17:49,180 --> 00:17:52,080
I know that solution architect means
different things at different places.

254
00:17:52,120 --> 00:17:55,200
At Crunchy Data, it was a
hybrid role, so you did some

255
00:17:56,100 --> 00:18:01,720
Postgres consultancy, some
joining in on pre-sales calls.

256
00:18:03,160 --> 00:18:04,040
A lot of

257
00:18:07,440 --> 00:18:14,700
kind of evangelism, I suppose, going out and
talking about Postgres and meeting developers and,

258
00:18:16,360 --> 00:18:17,800
yeah, it was great because

259
00:18:18,500 --> 00:18:23,840
I just got to talk about Postgres, play with
Postgres, and teach other people about Postgres.

260
00:18:26,160 --> 00:18:26,600
That was great.

261
00:18:28,476 --> 00:18:35,876
[CLAIRE] And when you were wearing that role of
helping other DBAs, which in the Postgres consultancy

262
00:18:37,016 --> 00:18:37,256
role,

263
00:18:38,136 --> 00:18:40,476
you had said at the beginning
that you wanted to be a teacher.

264
00:18:40,836 --> 00:18:43,416
That kind of puts you in
teacher mode, didn't it? I mean,

265
00:18:44,136 --> 00:18:46,656
did it end up scratching two itches at once, or?

266
00:18:47,220 --> 00:18:51,320
[KAREN] Yes. Exactly that, because it's not
that I don't like teaching. I do like teaching.

267
00:18:51,560 --> 00:18:56,540
I just saw what being in the
education system done to my family and

268
00:18:57,200 --> 00:19:00,060
I didn't really want to do that. I love
teaching people that want to learn.

269
00:19:00,960 --> 00:19:03,860
I love teaching things that I'm excited about.

270
00:19:05,400 --> 00:19:07,820
So yeah, not having a curriculum, but

271
00:19:08,520 --> 00:19:14,980
being able to investigate new things or write
about things that I've discovered or things that

272
00:19:15,040 --> 00:19:16,840
customers have asked me, that kind of thing. And,

273
00:19:18,360 --> 00:19:23,220
yeah, create talks and training sessions and that
kind of thing. It's perfect. So, yeah, absolutely.

274
00:19:25,776 --> 00:19:29,536
[CLAIRE] At Crunchy, when you had
this hybrid solution architect role,

275
00:19:30,396 --> 00:19:31,856
did you find it was hard to,

276
00:19:32,796 --> 00:19:37,256
I don't know, manage your time across
those different areas of responsibility.

277
00:19:38,376 --> 00:19:39,796
Was that a challenge, or

278
00:19:40,476 --> 00:19:45,776
was it explicitly not a challenge because you were
just reacting to whatever the customers needed

279
00:19:45,936 --> 00:19:48,196
or whatever the team needed?

280
00:19:49,376 --> 00:19:51,656
I'm just always curious about time management.

281
00:19:52,720 --> 00:19:58,320
[KAREN] Yeah, time management is something
that's still a work in progress for me. So

282
00:19:58,900 --> 00:20:03,740
yes, it was reacting to what
customers needed at the time, but also

283
00:20:04,200 --> 00:20:07,940
things that I wanted to do. So

284
00:20:08,460 --> 00:20:11,880
my time management was just say yes
to it and then find the time to do

285
00:20:11,940 --> 00:20:12,040
it,

286
00:20:13,540 --> 00:20:15,480
which isn't necessarily a healthy approach.

287
00:20:16,756 --> 00:20:21,556
[CLAIRE] Did that mean some late nights or
some weekends or some early mornings at times?

288
00:20:22,120 --> 00:20:23,960
[KAREN] Oh, not early
mornings. No, no, no, no, no.

289
00:20:24,276 --> 00:20:25,556
[CLAIRE] Oh, not a morning person?

290
00:20:26,120 --> 00:20:27,240
[KAREN] Not a morning person.

291
00:20:27,756 --> 00:20:31,936
[CLAIRE] OK, so we're doing this recording
and you and I are not in the same time

292
00:20:32,016 --> 00:20:35,476
zone. So it's morning for
me, but it's evening for you.

293
00:20:36,756 --> 00:20:39,416
And you're saying you're perfectly
happy doing this in the evening.

294
00:20:39,636 --> 00:20:41,976
You wouldn't have wanted
to do it bright and early.

295
00:20:41,500 --> 00:20:45,560
[KAREN] No. Yes. This is when my
brain kind of seems to come to life

296
00:20:46,500 --> 00:20:48,040
and be happy to do things, yeah.

297
00:20:48,936 --> 00:20:54,136
[CLAIRE] Got it. Well, I have an espresso
in hand, so that that helps me function

298
00:20:53,260 --> 00:20:54,240
[KAREN] I mean, that works too.

299
00:20:54,896 --> 00:20:55,836
[CLAIRE] at this time of day.

300
00:20:59,796 --> 00:21:05,796
The title of this episode is From
Oracle DBA to Postgres Contributor, and

301
00:21:06,256 --> 00:21:08,796
I think the thing that I want to get to

302
00:21:09,796 --> 00:21:14,916
is you obviously moved from Oracle to Postgres
not through your own choice, but because it

303
00:21:14,956 --> 00:21:21,616
was forced on you. Hopefully you're ultimately
happy with that forced choice, but you were

304
00:21:22,796 --> 00:21:27,576
still operating as a Postgres DBA,
as a Postgres expert consultant,

305
00:21:28,056 --> 00:21:32,996
as a Postgres solution architect, at what
point did you get involved in the community

306
00:21:33,556 --> 00:21:39,236
and in contributing to the open
source project? How did that happen?

307
00:21:39,936 --> 00:21:44,856
Was there a single moment, a single
phone call or a single conference?

308
00:21:48,520 --> 00:21:55,920
[KAREN] So I think I was very, very
lucky because I went to PGDay France

309
00:21:56,540 --> 00:21:58,620
in 2019, which was in Lyon,

310
00:21:59,600 --> 00:22:01,740
which was where I was working at the time.

311
00:22:02,296 --> 00:22:04,956
[CLAIRE] I should point out that you speak French,

312
00:22:05,676 --> 00:22:09,516
like extremely good French, which is
good because you live in France, so.

313
00:22:06,080 --> 00:22:06,940
[KAREN] I do, yes.

314
00:22:09,380 --> 00:22:11,841
I live in France, so yeah, there's
not really an excuse not to.

315
00:22:12,676 --> 00:22:17,456
[CLAIRE] Because PGDay France is an all
French-speaking conference, is that right?

316
00:22:16,820 --> 00:22:18,500
[KAREN] It's a French-speaking conference, yep.

317
00:22:19,400 --> 00:22:19,580
Yep.

318
00:22:21,240 --> 00:22:24,000
So yeah, it was in Lyon that year,
which is where I was working.

319
00:22:24,340 --> 00:22:29,020
So I was able to just go along, and
some of my colleagues went as well,

320
00:22:30,360 --> 00:22:33,520
because we were all trying to figure out
what this Postgres thing was and how we

321
00:22:33,559 --> 00:22:34,300
were going to work with it.

322
00:22:35,540 --> 00:22:41,980
And, excuse me, and Lætitia was one
of the speakers there, Lætitia Avrot.

323
00:22:42,680 --> 00:22:43,920
So I

324
00:22:44,760 --> 00:22:46,560
only said a couple of words to her then.

325
00:22:48,600 --> 00:22:52,280
But I was able to go to PGConf.EU that year.

326
00:22:53,260 --> 00:22:57,200
And I met Lætitia again and said, Oh,
I don't know if you remember me, but

327
00:22:57,860 --> 00:23:00,240
we met in Lyon, and she

328
00:23:01,780 --> 00:23:07,040
introduced me then to, I can't even remember
who at the time because names and faces

329
00:23:07,080 --> 00:23:09,360
for me are a blur at the best of times.

330
00:23:09,820 --> 00:23:15,920
But she introduced me to so many people in the
Postgres community that during that conference.

331
00:23:17,736 --> 00:23:20,716
[CLAIRE] Can you remind me which
city this PGConf.EU was in?

332
00:23:21,760 --> 00:23:22,840
[KAREN] Milan, I think.

333
00:23:23,536 --> 00:23:26,196
[CLAIRE] Okay, so that would have been 2019.

334
00:23:26,880 --> 00:23:26,980
[KAREN] Yeah.

335
00:23:28,696 --> 00:23:29,036
[CLAIRE] Awesome.

336
00:23:28,760 --> 00:23:31,380
[KAREN] Yeah. Yeah, so that was amazing,

337
00:23:33,420 --> 00:23:36,880
to have somebody just say, oh, you know,
you need to talk to this person and

338
00:23:36,920 --> 00:23:38,120
this person and this person.

339
00:23:39,920 --> 00:23:44,320
So I came away from that feeling as though
I was part of the Postgres community,

340
00:23:44,400 --> 00:23:45,500
which was amazing.

341
00:23:49,256 --> 00:23:53,256
[CLAIRE] But for some people,
that would be like you would now

342
00:23:55,596 --> 00:24:00,256
have friends, have relationships,
have stronger acquaintances,

343
00:24:01,056 --> 00:24:05,636
and you would just come to PGConf.EU every year.
You wouldn't necessarily turn into a contributor.

344
00:24:06,136 --> 00:24:11,756
So something somewhere pushed
you into making contributions.

345
00:24:11,816 --> 00:24:14,976
You started giving a lot of conference
talks at Postgres conferences,

346
00:24:15,596 --> 00:24:20,336
sharing the expertise that you'd
built. And I don't, I'm just curious.

347
00:24:20,416 --> 00:24:21,236
I want to know more.

348
00:24:21,936 --> 00:24:22,736
How did that happen?

349
00:24:22,860 --> 00:24:24,320
[KAREN] It's a really good question.

350
00:24:24,436 --> 00:24:28,316
[CLAIRE] By the way, I was there in
Milan. I did not meet you, that at least.

351
00:24:29,836 --> 00:24:31,936
I don't remember meeting you in Milan. Do you?

352
00:24:33,400 --> 00:24:35,500
[KAREN] I don't remember, no. No.

353
00:24:33,680 --> 00:24:33,900
[CLAIRE] Uh-uh.

354
00:24:35,160 --> 00:24:35,960
Oh, good. I'm glad.

355
00:24:36,620 --> 00:24:37,640
That would have been embarrassing.

356
00:24:38,360 --> 00:24:41,260
[KAREN] Because I think the
next place I went was FOSDEM.

357
00:24:42,940 --> 00:24:46,940
2019, 2020, that must have been, yeah,
just before everything locked down.

358
00:24:47,480 --> 00:24:48,600
[CLAIRE] And I was there too.

359
00:24:49,740 --> 00:24:50,660
But did we meet?

360
00:24:51,780 --> 00:24:53,200
[KAREN] I think we did meet then.

361
00:24:53,920 --> 00:24:55,380
[CLAIRE] No, we met at FOSDEM.

362
00:24:56,360 --> 00:24:58,420
[KAREN] I remember, yeah, bumping into you

363
00:24:59,300 --> 00:25:05,320
walking between two buildings at some point,
but yeah. Yeah, how did I actually get involved?

364
00:25:05,700 --> 00:25:07,340
I think some of it was that.

365
00:25:08,280 --> 00:25:09,440
I kind of came to that

366
00:25:10,540 --> 00:25:14,500
realization, because with Oracle I was just

367
00:25:15,080 --> 00:25:18,540
a user of the product, and all of a
sudden I realized that there was this

368
00:25:18,600 --> 00:25:22,380
whole kind of community that could not
just be a user of the product, but actually

369
00:25:22,500 --> 00:25:29,620
help to guide the product and the project
and make it bigger and better. And

370
00:25:30,720 --> 00:25:32,980
so there was some of that, there
was some of that kind of...

371
00:25:34,540 --> 00:25:38,040
Yeah, realization that actually
there is this whole group of people

372
00:25:38,920 --> 00:25:41,740
working on the same thing,
but I really can't remember.

373
00:25:44,340 --> 00:25:45,240
You know, when I

374
00:25:46,580 --> 00:25:48,660
started becoming involved. It can't have been much

375
00:25:49,340 --> 00:25:51,920
later than that because it's
been quite a few years now

376
00:25:53,640 --> 00:25:56,900
that I've been doing things
for. It's probably just because

377
00:25:59,100 --> 00:26:00,780
I can't say no. So

378
00:26:02,760 --> 00:26:06,240
if someone asked me to do something,
I would probably say, yeah, why not?

379
00:26:06,340 --> 00:26:07,400
I can help with that.

380
00:26:08,540 --> 00:26:10,820
[CLAIRE] Well, when I think
of all the different types of

381
00:26:11,380 --> 00:26:16,700
contributions you've made to the upstream
project, I'll enumerate some of them, and then

382
00:26:18,000 --> 00:26:20,980
I'm sure I will miss some because
I do not have a list in front of

383
00:26:21,020 --> 00:26:22,800
me, so this is all from memory.

384
00:26:23,200 --> 00:26:26,320
But I know that you served on the
board of PostgreSQL Europe for a while.

385
00:26:27,200 --> 00:26:31,840
I know that I served with you on the
PGDay Chicago talk selection team

386
00:26:32,320 --> 00:26:35,420
a number of several years ago. That must
have been at least three years ago now.

387
00:26:35,900 --> 00:26:43,500
You've served on program committees for PGConf.EU
and probably several other conferences as well,

388
00:26:43,619 --> 00:26:49,680
probably FOSDEM PGDay and FOSDEM,
and I'm certain there are more.

389
00:26:51,480 --> 00:26:53,300
You serve on the— actually, you lead the

390
00:26:53,360 --> 00:26:56,560
Diversity Committee. I did mention
that earlier for PostgreSQL Europe.

391
00:26:58,060 --> 00:27:04,800
Organizing committees like PGConf.EU as well. So
there's, I think, and then you've been a speaker.

392
00:27:05,280 --> 00:27:09,860
And what I think is so cool
about your conference talks is

393
00:27:10,580 --> 00:27:15,720
you speak not only at Postgres conferences,
where you would expect to have users, but you

394
00:27:15,820 --> 00:27:21,460
also speak at app dev conferences like
DjangoCon, or I'm not sure if you've ever done

395
00:27:21,500 --> 00:27:25,260
a PyCon, but I know you've
keynoted DjangoCon one year.

396
00:27:26,260 --> 00:27:29,440
[KAREN] I did. That was a really proud moment.

397
00:27:31,540 --> 00:27:36,400
[CLAIRE] So, and I think it's fabulous to
get people in the app developer community

398
00:27:37,560 --> 00:27:42,740
who sometimes are a little bit far removed
from the underlying database, right, with the

399
00:27:44,100 --> 00:27:47,120
ORM stuck in between, and maybe
they don't fully understand it.

400
00:27:47,200 --> 00:27:51,400
Maybe they want to understand more,
but they've never gotten to that point.

401
00:27:51,800 --> 00:27:57,120
And so anyway, I love that you're also
teaching people in those communities as well.

402
00:27:58,180 --> 00:28:01,060
[KAREN] Yeah. I really, really love
going into those communities. Yeah, so I

403
00:28:02,220 --> 00:28:05,140
went to my first DjangoCon in 2022.

404
00:28:06,920 --> 00:28:11,980
Yeah, and that was actually Craig Kerstiens
who suggested that I might want to

405
00:28:12,820 --> 00:28:13,640
submit a talk

406
00:28:14,420 --> 00:28:17,140
and go and speak to the folks at DjangoCon.

407
00:28:17,980 --> 00:28:21,860
And so I did that in 2022, and I
was really excited to get a talk

408
00:28:21,940 --> 00:28:27,080
accepted. And I enjoyed that
conference so much. I enjoyed the—

409
00:28:28,700 --> 00:28:33,560
just the feeling from the audience, the questions
afterwards, the interactions with everybody,

410
00:28:34,200 --> 00:28:35,200
and I've actually

411
00:28:36,140 --> 00:28:37,820
managed to go to DjangoCon,

412
00:28:38,980 --> 00:28:44,400
apart from this year, every year since
then. So I did 2022 through to 2025.

413
00:28:46,840 --> 00:28:53,220
Because they just, they love Postgres. The Django
community is so enthusiastic about Postgres, and

414
00:28:54,240 --> 00:28:56,460
they want to learn about it. So yeah, you said

415
00:28:57,160 --> 00:29:01,260
they're often quite removed from the database,
but they want to know what's going on.

416
00:29:02,920 --> 00:29:05,360
So it's been really fun to write talks

417
00:29:06,300 --> 00:29:09,840
specifically for them to help them understand
the things that are useful to them.

418
00:29:11,100 --> 00:29:16,260
[CLAIRE] Yeah, when I worked at Citus Data,
which is a Postgres startup that got acquired

419
00:29:16,920 --> 00:29:21,140
by Microsoft back in 2019, and I
worked with Craig Kerstiens as well,

420
00:29:22,080 --> 00:29:24,340
we used to go every year to PyCon

421
00:29:25,120 --> 00:29:27,660
and have a booth and answer people's questions.

422
00:29:27,800 --> 00:29:33,960
And the affinity, the love, the affection, the
interest that Python developers have for Postgres

423
00:29:34,060 --> 00:29:40,180
just overwhelm me every single time. We would
have standing room only at our booth and

424
00:29:41,620 --> 00:29:47,160
deep lines of people wanting to come and
talk and discuss Postgres questions, and

425
00:29:48,000 --> 00:29:48,920
it was pretty cool.

426
00:29:49,840 --> 00:29:52,040
[KAREN] Yeah, I love it. It,
yeah, it warms the heart.

427
00:29:53,720 --> 00:29:59,240
And then I went to a couple of Ruby
conferences as well over the last

428
00:30:00,060 --> 00:30:03,160
couple of years, and I got that
same feeling there as well.

429
00:30:03,280 --> 00:30:05,100
It was just so lovely, yeah.

430
00:30:10,700 --> 00:30:18,320
[CLAIRE] So, PGConf.EU, we just talked about you serving
on the organizing team and past program committees.

431
00:30:18,480 --> 00:30:21,100
I don't know if you're on the
program committee this year or not,

432
00:30:21,780 --> 00:30:27,280
but I know you have been in the past. It's
happening next week in Valencia, Spain.

433
00:30:28,280 --> 00:30:33,540
So for those of you listening, if you're not
familiar, PGConf.EU is the largest in-person Postgres

434
00:30:33,640 --> 00:30:38,600
conference of the year, and it rotates around
different countries and cities in Europe.

435
00:30:39,200 --> 00:30:44,000
So every year it's in a different location,
and there's always a bit of anticipation

436
00:30:44,480 --> 00:30:47,820
in terms of where's it going to be
next year. They try to organize it

437
00:30:48,516 --> 00:30:49,616
a year in advance.

438
00:30:49,696 --> 00:30:53,556
Sometimes it's only 11 months in advance,
but as soon as they get the contract signed.

439
00:30:54,076 --> 00:30:58,736
And so this year it's in Spain, so
it should be warmer than it was last

440
00:30:58,776 --> 00:31:06,396
year in Riga. And are you, I know
you're giving a talk. I looked it up.

441
00:31:00,750 --> 00:31:01,070
[KAREN] Yes.

442
00:31:06,516 --> 00:31:09,496
[CLAIRE] You're giving a talk on
tuning Postgres for analytics.

443
00:31:11,536 --> 00:31:15,076
And I will do my darndest to be there.

444
00:31:15,356 --> 00:31:19,876
Looks like Tuesday at five o'clock,
so I am going to try to be there,

445
00:31:20,876 --> 00:31:22,236
see what you've been up to lately.

446
00:31:22,890 --> 00:31:26,570
[KAREN] Just before the social event, so
hopefully everybody will want to be around.

447
00:31:27,836 --> 00:31:28,336
[CLAIRE] There you go.

448
00:31:30,196 --> 00:31:34,996
Is there anything in particular that you
are super looking forward to next week?

449
00:31:36,596 --> 00:31:40,296
Anything I need to pay attention to.

450
00:31:39,190 --> 00:31:42,390
[KAREN] Do you know what? I
am so embarrassed to say this.

451
00:31:42,690 --> 00:31:46,830
I have not properly looked at the schedule
this year, at the talks on the schedule,

452
00:31:47,430 --> 00:31:48,070
because

453
00:31:50,770 --> 00:31:54,830
I was really glad I had Samed helping me on the

454
00:31:55,810 --> 00:31:57,170
program committee last year.

455
00:31:57,930 --> 00:32:02,670
So I was the chair and he was helping
me, and he did such an amazing

456
00:32:02,730 --> 00:32:02,990
job

457
00:32:03,710 --> 00:32:06,870
that I suggested delegating it to him this year.

458
00:32:07,110 --> 00:32:10,670
So he's taken it on this year, and
he's done, from what I can see, an

459
00:32:10,730 --> 00:32:11,410
amazing job.

460
00:32:12,830 --> 00:32:16,230
And I do remember having a quick look
through to see who was on the schedule

461
00:32:16,270 --> 00:32:17,170
and what was going on.

462
00:32:17,830 --> 00:32:23,550
But I haven't, since then, I've been so busy
organizing PGConf.EU that I haven't actually sat

463
00:32:23,610 --> 00:32:27,010
down to work out which talks I
want to go and see when I'm there.

464
00:32:27,436 --> 00:32:29,196
[CLAIRE] So you don't have a
plan yet, is what you're saying.

465
00:32:30,310 --> 00:32:31,130
[KAREN] Yeah, pretty much, yeah.

466
00:32:31,656 --> 00:32:35,356
[CLAIRE] Okay, that's fine. I mean,
there's two ways to do it, right? One way

467
00:32:36,016 --> 00:32:40,436
you show up with your calendar pre-booked or
a spreadsheet that tells you what, where to

468
00:32:40,476 --> 00:32:45,236
go when, or you just at the start of
every session, you just look at your

469
00:32:45,256 --> 00:32:48,956
phone and you roll the dice,
and that works too. It's,

470
00:32:49,456 --> 00:32:53,476
I will say that at PGConf.EU, unlike FOSDEM, where

471
00:32:54,396 --> 00:32:59,676
you sometimes have to get in line in
the queue for a particular talk up to

472
00:32:59,716 --> 00:33:03,476
an hour before, or you will
not get a seat in that talk.

473
00:33:04,076 --> 00:33:09,056
PGConf.EU, you can almost always squeeze
into a room, which is not to say the rooms

474
00:33:09,096 --> 00:33:12,476
aren't full. It's not to say
it's not a large conference, but

475
00:33:13,176 --> 00:33:14,616
the rooms are big enough.

476
00:33:15,316 --> 00:33:17,636
You know, and there's standing
room in the back usually.

477
00:33:17,896 --> 00:33:21,936
FOSDEM, because it's on a university
campus, those rooms are small,

478
00:33:22,396 --> 00:33:25,076
and they're very strict about capacity limits, so.

479
00:33:27,250 --> 00:33:32,990
[KAREN] Yeah, no, I like to go to every
conference with a full plan of exactly where

480
00:33:33,030 --> 00:33:36,790
I'm going to be where, but this next
week isn't working out like that, so.

481
00:33:37,516 --> 00:33:39,516
[CLAIRE] But you still have a couple of days.

482
00:33:40,130 --> 00:33:41,690
[KAREN] I do have a couple of days, yeah.

483
00:33:41,810 --> 00:33:44,350
But there are so many other things
that I want to do as well, and so

484
00:33:44,430 --> 00:33:47,730
many people that I haven't seen this
year that I want to talk to, and

485
00:33:49,370 --> 00:33:50,290
so I'm going to have to be

486
00:33:51,310 --> 00:33:52,610
quite selective, I think,

487
00:33:54,270 --> 00:33:55,470
about which talks I go to.

488
00:33:54,636 --> 00:34:00,376
[CLAIRE] Okay, so we should talk about the
elephant in the room, and I'm not talking about

489
00:34:00,416 --> 00:34:06,216
a Postgres elephant, although immediately as I
said that, I'm picturing a Postgres elephant.

490
00:34:07,076 --> 00:34:11,256
Oh, here, I'm holding one up. Nobody can
see it because this is an audio podcast.

491
00:34:12,316 --> 00:34:17,716
You were originally supposed to be on this
podcast at the end of last calendar year,

492
00:34:17,776 --> 00:34:19,476
at the end of 2025.

493
00:34:20,290 --> 00:34:22,270
[KAREN] No. I was, and I know exactly which date.

494
00:34:24,776 --> 00:34:26,876
[CLAIRE] We had the sound check done.

495
00:34:27,876 --> 00:34:33,016
Everything was set, graphics were
ready, we were prepared to go, and then

496
00:34:33,856 --> 00:34:34,636
you had to cancel.

497
00:34:36,690 --> 00:34:41,670
[KAREN] Yes. Yeah, that was
a bit of a shock last year.

498
00:34:41,770 --> 00:34:44,970
I remember I had been in touch with
you, I think two days prior to say,

499
00:34:45,030 --> 00:34:47,510
Claire, I don't know how
this is going to go because

500
00:34:48,750 --> 00:34:50,590
things are looking a bit weird.

501
00:34:51,590 --> 00:34:55,050
And then yes, the day that we were
supposed to record was the day that my

502
00:34:55,070 --> 00:34:56,690
husband was diagnosed with leukemia.

503
00:34:58,570 --> 00:35:02,770
So since then, he's okay at the moment,
so I will say that straight away before

504
00:35:02,810 --> 00:35:03,911
I go into the rest of the story:

505
00:35:05,810 --> 00:35:08,110
he is in remission. The cancer's in remission.

506
00:35:09,470 --> 00:35:14,110
He's finished his intensive treatment, and we're
now kind of moving on to the maintenance phase.

507
00:35:14,150 --> 00:35:18,530
So, yes, we get that bit over with
first before I say anything else.

508
00:35:20,270 --> 00:35:27,990
And I can talk about it now without getting
emotional, but yeah, it was for months.

509
00:35:28,070 --> 00:35:29,650
It's been, it has been really tough.

510
00:35:30,730 --> 00:35:34,270
So yeah, 10 months of intensive treatment.

511
00:35:35,296 --> 00:35:38,636
[CLAIRE] And I know you had to step
back from a lot of things and just focus

512
00:35:38,696 --> 00:35:42,076
on supporting your husband and supporting
the family, which was, I think,

513
00:35:43,176 --> 00:35:47,536
a big burden to bear. I think that's probably
why you stepped back from the Postgres

514
00:35:48,216 --> 00:35:49,776
nonprofit board in Europe.

515
00:35:50,030 --> 00:35:50,910
[KAREN] Yes. Yep.

516
00:35:51,016 --> 00:35:53,516
[CLAIRE] I'm assuming that
Snowflake was supportive.

517
00:35:54,510 --> 00:35:56,410
[KAREN] Extremely supportive, yeah.

518
00:36:00,276 --> 00:36:03,056
[CLAIRE] I know that I, we were
going to see each other at FOSDEM.

519
00:36:03,576 --> 00:36:08,816
We were going to be celebrating at FOSDEM.
And obviously you didn't come to that

520
00:36:04,170 --> 00:36:05,510
[KAREN] I was really excited.

521
00:36:09,516 --> 00:36:15,176
[CLAIRE] conference either and just
hunkered down to help your husband.

522
00:36:14,790 --> 00:36:20,670
[KAREN] Yeah, very much. So I resigned
from the PostgreSQL Europe board.

523
00:36:21,730 --> 00:36:23,630
I think in January.

524
00:36:25,890 --> 00:36:31,850
After hiding away, the first
five weeks my husband was in a

525
00:36:32,330 --> 00:36:36,110
sterile room, so we could visit, but
only two people could visit at a time and

526
00:36:36,150 --> 00:36:40,350
only for a certain amount of time, and
it was, you know, behind so many...

527
00:36:41,182 --> 00:36:44,662
Buzzer, you know, doors with buzzers
and all that kind of thing. It was,

528
00:36:45,362 --> 00:36:46,342
yeah, it was quite stressful.

529
00:36:48,342 --> 00:36:50,402
And so he was in there over
Christmas and New Year.

530
00:36:50,602 --> 00:36:56,022
So for the first five weeks, I
just, I didn't do anything else.

531
00:36:56,122 --> 00:36:57,682
I didn't speak to anybody else really.

532
00:36:58,722 --> 00:37:03,042
And then I gradually managed to get back
to work, and I needed to work because

533
00:37:03,142 --> 00:37:06,502
I needed something to focus on. I
needed some kind of structure to my day.

534
00:37:08,322 --> 00:37:11,842
But I didn't travel because
I needed to be close to home.

535
00:37:13,382 --> 00:37:16,402
You know, going backwards and forwards
to the hospital and that kind of thing.

536
00:37:21,402 --> 00:37:23,802
I think I looked back earlier at doing kind of

537
00:37:24,582 --> 00:37:27,342
ten, twelve, sometimes even more

538
00:37:28,202 --> 00:37:31,542
trips to do conference talks each year. This year,

539
00:37:33,202 --> 00:37:38,862
I did go to EuroPython in Krakow
in July, which was amazing because

540
00:37:40,142 --> 00:37:44,102
EuroPython is one of the conferences
that I had wanted to go to for ages, and

541
00:37:44,182 --> 00:37:49,122
Mike was well and he was in between
treatments, so I was able to go, which

542
00:37:49,202 --> 00:37:56,562
was brilliant. But other than that, PGConf.EU
next week will be my first conference.

543
00:37:56,562 --> 00:37:56,567
It's been an interesting year.

544
00:38:01,786 --> 00:38:06,386
[CLAIRE] I am very, very happy for you
and your family that he's in remission.

545
00:38:09,386 --> 00:38:10,746
And that we're having this conversation

546
00:38:11,986 --> 00:38:12,646
instead of...

547
00:38:12,482 --> 00:38:12,862
[KAREN] Yes.

548
00:38:13,942 --> 00:38:15,202
Yes, absolutely.

549
00:38:16,186 --> 00:38:18,766
[CLAIRE] And I get to see you
in person next week, which is,

550
00:38:20,126 --> 00:38:25,506
I'm, I know a lot of people are gonna
be, because you haven't been as visible

551
00:38:25,966 --> 00:38:30,106
and as present in the community in
the last year, you're probably gonna,

552
00:38:30,806 --> 00:38:34,746
get to your hotel room at night and just
be exhausted because you're going to get

553
00:38:34,826 --> 00:38:38,846
pulled in a million directions. We're all
going to want a little bit of Karen time,

554
00:38:40,026 --> 00:38:40,666
and so...

555
00:38:41,666 --> 00:38:41,866
Ah.

556
00:38:42,726 --> 00:38:43,586
Hopefully I can.

557
00:38:43,322 --> 00:38:43,722
[KAREN] No, I am.

558
00:38:45,722 --> 00:38:47,422
I am really looking forward to seeing everyone.

559
00:38:48,186 --> 00:38:50,166
[CLAIRE] Or buy you a tea or something like that.

560
00:38:50,362 --> 00:38:50,521
[KAREN] Yes.

561
00:38:52,166 --> 00:38:58,926
[CLAIRE] Okay, so PGConf.EU is expected to
be bigger this year than it was last year.

562
00:38:59,406 --> 00:39:00,326
Is that something?

563
00:39:00,082 --> 00:39:00,482
[KAREN] Yes.

564
00:39:02,726 --> 00:39:06,066
[CLAIRE] And for the very first
time, it's not being held in a hotel.

565
00:39:06,086 --> 00:39:12,486
Because there's this really interesting thing, like
these price points, you know what I mean? Hotels

566
00:39:13,446 --> 00:39:16,186
can only get so big in terms
of the size of the conference.

567
00:39:16,526 --> 00:39:21,846
But then after hotels, what's the
next step? And is it too much bigger?

568
00:39:22,166 --> 00:39:26,026
Is there an in-between? I don't even know
how you found this conference center, but...

569
00:39:27,506 --> 00:39:28,406
It'll be different.

570
00:39:29,562 --> 00:39:32,582
[KAREN] It will be different. I think

571
00:39:34,222 --> 00:39:39,982
there's enough of the kind of usual PGConf.EU
vibe that it'll still feel familiar to people.

572
00:39:41,582 --> 00:39:46,402
And there is the hotel just kind of next
door to the conference center, so hopefully

573
00:39:46,462 --> 00:39:48,402
that will kind of feel a bit like an extension

574
00:39:49,562 --> 00:39:54,042
of the conference center. So
yeah, but it's a new experience.

575
00:39:54,102 --> 00:39:59,782
It's been a huge learning curve for the
organization team, of course, because

576
00:40:00,902 --> 00:40:03,122
you've got your checklists for how things work

577
00:40:03,962 --> 00:40:07,342
in a hotel venue, and obviously the checklists
are all different, aren't they, once you move

578
00:40:07,382 --> 00:40:09,442
on to a different kind of venue.

579
00:40:11,662 --> 00:40:14,382
But yeah, I'm excited about
it. I think it'll be good.

580
00:40:17,246 --> 00:40:21,946
[CLAIRE] Now that your husband is doing better, you're
coming to PGConf.EU, you're giving a talk there.

581
00:40:22,486 --> 00:40:27,566
Are you also submitting to other CFPs that
are open now? Is that something that you're

582
00:40:28,126 --> 00:40:31,746
ready to do, or are you going to be
more selective this year and only submit

583
00:40:31,786 --> 00:40:34,546
to a few places? Have you decided yet?

584
00:40:38,042 --> 00:40:40,682
[KAREN] I've realized that although
I've missed people this year,

585
00:40:41,262 --> 00:40:43,662
and I want to do more than I have this year,

586
00:40:44,702 --> 00:40:49,942
it's not sustainable for me to go back
to traveling more than once a month.

587
00:40:50,602 --> 00:40:55,242
And it's made me realize that I don't
feel as though I need that level of,

588
00:40:57,542 --> 00:41:00,782
you know, travel and that kind of thing. So

589
00:41:01,302 --> 00:41:05,702
I think I'm going to try and do maybe four to six

590
00:41:06,762 --> 00:41:07,922
conferences next year.

591
00:41:09,942 --> 00:41:12,982
And just, yeah, be selective about the ones I do.

592
00:41:13,162 --> 00:41:17,542
I would love to go back to
DjangoCon and EuroPython if I can.

593
00:41:20,382 --> 00:41:22,942
Yeah, I just, I really love bringing

594
00:41:23,842 --> 00:41:25,522
more Postgres knowledge to those

595
00:41:26,462 --> 00:41:31,882
communities and selfishly, I just love
interacting with them. I love being there.

596
00:41:33,562 --> 00:41:36,682
So they would be good. PGConf.EU is a must.

597
00:41:37,862 --> 00:41:40,462
And aside from that, I haven't decided yet.

598
00:41:38,286 --> 00:41:38,786
[CLAIRE] Of course.

599
00:41:41,286 --> 00:41:48,846
Okay. Well, I know that FOSDEM PGDay, so
FOSDEM PGDay is technically a fringe event.

600
00:41:49,826 --> 00:41:54,646
Around the weekend of FOSDEM, there are lots
of events, like the Thursday and Friday before,

601
00:41:54,686 --> 00:41:59,846
or the Monday and Tuesday after, and
FOSDEM PGDay always runs back to back

602
00:42:00,546 --> 00:42:04,026
in Brussels with FOSDEM itself, but the
exciting thing this year is it's going to be

603
00:42:04,086 --> 00:42:09,886
a two-day pre-event, and I don't
think the CFP is open for that yet.

604
00:42:09,926 --> 00:42:15,266
And then, of course, hopefully there will
be a Postgres devroom at FOSDEM this year.

605
00:42:15,346 --> 00:42:18,366
There was not last year, but
hopefully we'll find out.

606
00:42:18,506 --> 00:42:21,266
I don't think the decisions
have come out yet on that.

607
00:42:21,822 --> 00:42:24,262
[KAREN] I think the call for devrooms and

608
00:42:25,082 --> 00:42:26,562
booths is still open at the moment.

609
00:42:27,006 --> 00:42:32,406
[CLAIRE] Okay, cool. So I know that
that's on the table. PGConf.dev

610
00:42:33,006 --> 00:42:38,366
is happening in Montreal this
year in May, and that is, I'm not

611
00:42:38,926 --> 00:42:43,926
telling people where Karen is going to submit.
We didn't pre-discuss this. I'm not giving hints.

612
00:42:44,046 --> 00:42:47,986
I'm just talking about CFPs
that I'm excited about.

613
00:42:49,206 --> 00:42:55,426
And then PGConf India's CFP closes tomorrow, and
I'm still trying to decide whether to submit.

614
00:42:56,626 --> 00:43:01,146
I've been two years in a row, and it's
got—it's a really wonderful conference with a

615
00:43:01,206 --> 00:43:07,246
lot of energy and a lot of excitement
around learning more about Postgres.

616
00:43:08,106 --> 00:43:11,986
It's just a very long flight
for me, so I haven't decided yet

617
00:43:12,066 --> 00:43:13,526
if I could, you know what I mean?

618
00:43:13,586 --> 00:43:16,306
If I get accepted, can I handle that flight again?

619
00:43:18,756 --> 00:43:21,196
And then we are doing POSETTE again this year.

620
00:43:21,796 --> 00:43:26,336
This is going to be the sixth year,
so POSETTE: An Event for Postgres,

621
00:43:27,056 --> 00:43:30,276
and the save the date just
went out, and it's on the 30th

622
00:43:30,916 --> 00:43:32,076
and 31st of March

623
00:43:32,996 --> 00:43:38,556
next year. So those are the CFPs
that are on my mind, at least.

624
00:43:44,016 --> 00:43:47,316
We talked for a second about the
keynote you gave at DjangoCon

625
00:43:48,216 --> 00:43:51,836
a couple of years ago, and I'll
make sure to include a link to that

626
00:43:52,536 --> 00:43:55,636
in the show notes. But I think the title was,

627
00:43:56,336 --> 00:44:02,276
When I grow up, I want to be a database
administrator, said no one ever.

628
00:44:05,036 --> 00:44:07,396
And I'm just curious what the
reception was like to that.

629
00:44:08,532 --> 00:44:13,792
[KAREN] It was really good, actually, and
more to the point, I really enjoyed giving it,

630
00:44:15,032 --> 00:44:21,192
because it allowed me to be a bit
self-indulgent and talk about how I

631
00:44:22,032 --> 00:44:24,752
became a database administrator,
and the fact that even though

632
00:44:26,772 --> 00:44:31,172
my dad taught me to count in
binary when I was a toddler,

633
00:44:32,392 --> 00:44:37,692
he bought me my first computer.
Well, he built me my first computer

634
00:44:38,332 --> 00:44:41,732
when I was, I think I must have been
about nine or something like that, and

635
00:44:41,792 --> 00:44:43,572
taught me to program in BASIC and...

636
00:44:45,912 --> 00:44:46,832
You know, I was

637
00:44:48,192 --> 00:44:54,432
obviously a bit of a geek, and I did a
maths degree and software development MSc.

638
00:44:56,312 --> 00:44:56,412
I

639
00:44:57,972 --> 00:45:00,871
didn't even consider being
a database administrator.

640
00:45:00,952 --> 00:45:03,312
I didn't know what a database administrator was.

641
00:45:05,352 --> 00:45:08,292
It's just not something
that's on most people's radar.

642
00:45:08,732 --> 00:45:13,572
And even if it is on people's radar, it's
not the kind of glamorous, exciting job

643
00:45:13,632 --> 00:45:15,072
that people dream of.

644
00:45:16,272 --> 00:45:20,052
So I just, I really like the idea
of doing a kind of tongue-in-cheek:

645
00:45:22,092 --> 00:45:25,572
you know, these are all the things that
I love about databases and about being a

646
00:45:25,612 --> 00:45:29,592
database administrator, and let me try
and share some of that love with you.

647
00:45:32,156 --> 00:45:35,776
[CLAIRE] I think it's so cool
that you took that aptitude test

648
00:45:36,576 --> 00:45:41,696
shortly out of university, and it ended
up spitting out that you should have this

649
00:45:42,216 --> 00:45:42,736
job type.

650
00:45:43,976 --> 00:45:49,676
I'm very surprised that an aptitude test
even focused on the niche world of being a

651
00:45:49,716 --> 00:45:55,496
database administrator, but clearly it's been
a good fit for you in those earlier years of

652
00:45:55,536 --> 00:45:55,976
your career.

653
00:45:56,952 --> 00:46:00,612
[KAREN] I feel so lucky that
they did those tests because,

654
00:46:02,912 --> 00:46:07,732
I don't know, I might never have discovered
databases otherwise, but clearly it was a good

655
00:46:03,036 --> 00:46:03,636
[CLAIRE] Who knows?

656
00:46:07,792 --> 00:46:12,752
[KAREN] fit for me because I wouldn't have carried
on being a database administrator for 20 years

657
00:46:12,792 --> 00:46:13,052
and then

658
00:46:14,212 --> 00:46:16,212
staying in the world of databases for

659
00:46:16,992 --> 00:46:18,272
however many years longer, yeah.

660
00:46:19,496 --> 00:46:23,016
[CLAIRE] And now that you're at Snowflake,
I imagine your role has changed.

661
00:46:22,112 --> 00:46:22,532
[KAREN] Yes.

662
00:46:23,576 --> 00:46:26,476
[CLAIRE] I can see that your job
title on LinkedIn has changed.

663
00:46:27,252 --> 00:46:27,812
[KAREN] It has.

664
00:46:28,276 --> 00:46:32,916
[CLAIRE] And you are now a senior software
engineer. So what is, what does that mean?

665
00:46:33,516 --> 00:46:35,416
How's your focus different now?

666
00:46:36,112 --> 00:46:39,872
[KAREN] So this is really
exciting for me, actually, because

667
00:46:42,392 --> 00:46:48,292
I've already said how much I loved the
solution architect role at Crunchy Data.

668
00:46:51,652 --> 00:46:56,912
But I was starting to feel as though I
wasn't getting enough time to be kind

669
00:46:56,972 --> 00:46:57,212
of

670
00:46:58,132 --> 00:46:59,072
properly technical.

671
00:47:00,072 --> 00:47:02,752
Because I was being pulled in
lots of different directions.

672
00:47:04,552 --> 00:47:07,713
And then the solution architect role at

673
00:47:10,652 --> 00:47:14,512
Snowflake is completely different, I think.

674
00:47:17,292 --> 00:47:23,592
The solution architect role at Snowflake
is more of a kind of consulting type role.

675
00:47:24,432 --> 00:47:30,412
And then the role that most of us were being
moved towards was solution engineering, which

676
00:47:30,472 --> 00:47:34,132
is more of a kind of pre-sales,
technical pre-sales kind of role.

677
00:47:35,392 --> 00:47:37,672
And I'm quite happy to do some

678
00:47:38,652 --> 00:47:40,772
pre-sales, but it's really not where my

679
00:47:41,552 --> 00:47:45,152
interest lies. It's not the kind of thing
that excites me. What really excites me is

680
00:47:47,992 --> 00:47:51,132
building things or helping to build things and

681
00:47:52,352 --> 00:47:56,332
playing with things and teaching
other people about them. And

682
00:47:56,792 --> 00:48:01,312
so I've been really fortunate that
I have moved over to the engineering

683
00:48:02,872 --> 00:48:08,832
organization into the Postgres
extensions team at Snowflake.

684
00:48:09,512 --> 00:48:14,932
So I'm working with the folks who create pg_lake.

685
00:48:17,412 --> 00:48:25,052
And I'm doing a kind of a hybrid role
there where I am doing some development.

686
00:48:25,332 --> 00:48:32,372
I'm helping to develop internal
tooling and also doing all of the

687
00:48:32,952 --> 00:48:36,632
advocacy and that kind of thing
that I was doing before. So

688
00:48:37,352 --> 00:48:40,312
kind of picking and choosing my favorite bits

689
00:48:41,472 --> 00:48:42,732
of all the jobs really, so,

690
00:48:44,852 --> 00:48:45,772
yeah, it's great.

691
00:48:48,576 --> 00:48:53,256
[CLAIRE] Well, I think one is lucky if
you land in a place where you get to

692
00:48:53,316 --> 00:48:56,676
do the favorite bits of your
job. Not everybody gets that, so.

693
00:48:58,340 --> 00:49:00,240
[KAREN] Yeah, no, I feel very fortunate.

694
00:49:03,376 --> 00:49:07,416
[CLAIRE] I mentioned POSETTE a second
ago, and as I did it, I typed in a quick

695
00:49:07,536 --> 00:49:11,136
search to remind myself of
your past POSETTE talks.

696
00:49:11,696 --> 00:49:14,636
You've given three POSETTE talks over the years.

697
00:49:15,160 --> 00:49:17,360
[KAREN] Oh, have I? I remember two.

698
00:49:17,960 --> 00:49:17,961
[CLAIRE] Yeah.

699
00:49:17,960 --> 00:49:23,320
Okay, so, well, okay, technically one of them
was Citus Con, but I consider, remember how

700
00:49:23,360 --> 00:49:27,720
we renamed Citus Con: An Event for Postgres
to be POSETTE: An Event for Postgres.

701
00:49:27,760 --> 00:49:29,500
So anyway, it's three.

702
00:49:30,100 --> 00:49:35,380
And one was about multi-tenancy, one was about
how you don't need a database backup policy,

703
00:49:35,840 --> 00:49:39,800
and one was about managing Postgres at
scale: challenges, tools, and techniques.

704
00:49:40,480 --> 00:49:41,400
So I want to include

705
00:49:42,060 --> 00:49:45,240
at least one of these in the
show notes for the episode.

706
00:49:45,900 --> 00:49:46,760
Do you have a favorite?

707
00:49:50,560 --> 00:49:52,720
You've given so many talks in so many places.

708
00:49:52,780 --> 00:49:54,660
How could you recall what the favorites are?

709
00:49:55,780 --> 00:49:58,560
[KAREN] Yeah, no, I do quite
like the one with cakes in.

710
00:50:02,280 --> 00:50:04,560
[CLAIRE] That, that was... wasn't that
'You don't need a database backup policy'?

711
00:50:05,100 --> 00:50:06,140
Is that the one with cakes?

712
00:50:07,140 --> 00:50:11,860
[KAREN] Cakes, I think,
was multi-tenant databases.

713
00:50:12,980 --> 00:50:13,520
[CLAIRE] Really?

714
00:50:13,020 --> 00:50:17,360
[KAREN] Because obviously you would
describe multi-tenant databases with cakes.

715
00:50:17,920 --> 00:50:20,240
But no, you don't need a database
backup policy, I quite like.

716
00:50:20,280 --> 00:50:22,880
I had an explosion in there, so it's all good.

717
00:50:23,860 --> 00:50:27,740
[CLAIRE] Yeah, I'm looking at the slides right
now. There are no cakes in the backup policy.

718
00:50:28,480 --> 00:50:31,480
Okay, so it's got to be
multi-tenancy. All right, well,

719
00:50:32,820 --> 00:50:35,880
I'll roll the dice and I'll include
one of them in the show notes in case

720
00:50:35,940 --> 00:50:39,900
anybody's curious. All right, so

721
00:50:41,080 --> 00:50:42,960
one of the things I'm really curious about

722
00:50:44,000 --> 00:50:44,840
is how

723
00:50:45,820 --> 00:50:46,320
AI

724
00:50:47,120 --> 00:50:52,560
and LLMs are changing your work, are
changing your workflows, are changing how you

725
00:50:53,200 --> 00:50:59,140
do your job. If you have any favorite tools
or favorite new things that you've tried,

726
00:51:01,100 --> 00:51:02,560
as a guest on this show,

727
00:51:03,680 --> 00:51:09,280
I had Simon Willison on a couple of times,
and he's kind of world-renowned in the

728
00:51:09,360 --> 00:51:11,560
space of AI and kind of on the cutting edge of

729
00:51:12,020 --> 00:51:16,240
all the latest models, not just from
the frontier companies, but also

730
00:51:17,480 --> 00:51:20,660
open weight models as well and
things coming out of China. And

731
00:51:21,380 --> 00:51:22,960
it's always a blast for me

732
00:51:23,560 --> 00:51:28,660
to kind of hear from him, but he's so
advanced, right, that I'm always pulling it

733
00:51:28,700 --> 00:51:34,760
back like, okay, okay, what should
the rest of us do? But I am curious

734
00:51:35,440 --> 00:51:36,800
how it's changing

735
00:51:37,840 --> 00:51:38,280
your job,

736
00:51:39,320 --> 00:51:42,020
and if we can compare notes.

737
00:51:42,980 --> 00:51:47,300
[KAREN] Yeah, I did listen to Simon
talking about typing in the shower

738
00:51:47,840 --> 00:51:49,680
because he suddenly had an idea to

739
00:51:50,960 --> 00:51:54,460
rewrite something about, and no,
I'm not quite doing that yet.

740
00:51:55,440 --> 00:51:57,660
I think that I'm probably around

741
00:51:58,280 --> 00:52:01,640
the place that a lot of people
are, which is trying to work out

742
00:52:02,500 --> 00:52:06,620
exactly where it fits in my
day-to-day. So I'm using it a lot.

743
00:52:06,800 --> 00:52:13,220
I'm using agents to help me,
particularly with creating my

744
00:52:14,020 --> 00:52:14,920
internal tooling.

745
00:52:16,600 --> 00:52:20,480
And I'm still trying to get that

746
00:52:21,240 --> 00:52:26,280
balance of how much do I delegate
to the agent and how much do I want

747
00:52:26,340 --> 00:52:27,720
to keep hold of myself.

748
00:52:29,280 --> 00:52:32,980
And I imagine that there are probably a
lot of other people in that situation.

749
00:52:33,080 --> 00:52:33,560
I'm not.

750
00:52:36,360 --> 00:52:40,440
I'm not comfortable at the moment
with just sending off a whole team of,

751
00:52:40,900 --> 00:52:44,480
you know, 20 agents to just kind
of get on with things on their own.

752
00:52:45,240 --> 00:52:46,740
I still very much like to

753
00:52:47,440 --> 00:52:49,120
handhold and make sure that I'm

754
00:52:50,160 --> 00:52:53,340
giving loads of feedback and reviewing things

755
00:52:55,620 --> 00:52:59,120
and also trying to make sure that I don't
forget things, that I don't forget how

756
00:52:59,180 --> 00:53:01,320
to do things, I don't forget how to...

757
00:53:04,340 --> 00:53:07,400
You know, how to design a spec for something, or

758
00:53:08,620 --> 00:53:10,460
not necessarily the nitty gritty

759
00:53:12,100 --> 00:53:17,219
details of coding, but yeah,
just the actual thinking and

760
00:53:18,440 --> 00:53:23,820
reviewing and going through
that kind of iterative process.

761
00:53:24,740 --> 00:53:25,220
So yeah, it's

762
00:53:26,520 --> 00:53:28,080
one of those things at the moment that I think

763
00:53:30,300 --> 00:53:34,180
I'm constantly thinking about
and constantly trying to work out

764
00:53:34,640 --> 00:53:37,180
what's the right balance
here, what's the right amount

765
00:53:38,480 --> 00:53:42,400
to be doing myself, what's the right
amount to be delegating to agents?

766
00:53:44,520 --> 00:53:50,040
[CLAIRE] So I, about three or four
weeks ago, hit an inflection point.

767
00:53:50,880 --> 00:53:56,860
And tried something new, and
I was absolutely blown away by

768
00:53:57,320 --> 00:53:59,520
how much more efficient.

769
00:54:00,580 --> 00:54:04,260
And it has made me and how it
has unblocked some of my work.

770
00:54:04,720 --> 00:54:06,200
So I'm just going to tell you what I did,

771
00:54:06,700 --> 00:54:10,300
and then you, if you're not already
doing it, you should totally try it.

772
00:54:10,600 --> 00:54:13,780
And for those of you who are listening
who are not already doing this, you should

773
00:54:13,840 --> 00:54:14,580
totally try it.

774
00:54:16,360 --> 00:54:20,560
When Adam Prout was on the show
back in, I think he was a guest in

775
00:54:20,660 --> 00:54:25,600
May, Adam is a distinguished engineer at
Microsoft. I almost said at Sun Microsystems.

776
00:54:25,680 --> 00:54:27,600
Wow, things, the old habits die hard.

777
00:54:27,880 --> 00:54:32,860
No, he's a distinguished engineer at Microsoft,
and he's the lead architect on Azure HorizonDB.

778
00:54:32,960 --> 00:54:36,860
And he was talking about how at that
point in time, and of course things have

779
00:54:36,920 --> 00:54:38,880
changed since May, but he

780
00:54:40,360 --> 00:54:45,720
thought of the agents as typists for him,
and that he didn't type code anymore, right?

781
00:54:45,820 --> 00:54:51,000
He used voice to interact. And I
haven't been using voice all this time.

782
00:54:51,960 --> 00:54:57,580
So I know that Wispr Flow is really popular.
I have another friend who strongly recommends

783
00:54:58,880 --> 00:55:03,540
FluidVoice because it runs locally on your
system, and so it's super fast and it's open

784
00:55:03,600 --> 00:55:04,040
source.

785
00:55:04,120 --> 00:55:08,460
And of course, you can just use the microphones
and the voice capabilities that come with

786
00:55:08,560 --> 00:55:11,260
frontier models. But I had not been using voice.

787
00:55:12,200 --> 00:55:13,500
And I started using it.

788
00:55:14,160 --> 00:55:17,936
And oh my gosh, it is so much faster than typing.

789
00:55:18,396 --> 00:55:22,456
Typing and backspacing and correcting
and typing and refining and rethinking.

790
00:55:23,076 --> 00:55:29,516
I literally will just ramble to the agents,
and it'll capture everything that I want.

791
00:55:29,796 --> 00:55:35,656
It pulls it out in an organized
manner and goes off and does the work.

792
00:55:36,216 --> 00:55:40,416
And so all of my instructions, all
of my prompts, all of my ideas are so

793
00:55:40,496 --> 00:55:43,496
much faster. And here's the
thing that blew me away.

794
00:55:44,716 --> 00:55:49,576
I had a particular proposal that I needed a
stakeholder to buy into inside of Microsoft,

795
00:55:50,056 --> 00:55:55,536
and I wanted to make sure that
the wording was particularly

796
00:55:56,016 --> 00:55:59,156
strong, powerful, influential,
persuasive. I wanted

797
00:55:59,196 --> 00:56:01,636
persuasive. I had the ideas in my head,

798
00:56:02,276 --> 00:56:04,856
and normally what I would
have done is I had these four

799
00:56:05,576 --> 00:56:09,596
persuasive points, I would have typed them
in, and then I would have actually sent it

800
00:56:09,656 --> 00:56:13,456
to an LLM to say, Is there anything
here that's ambiguous or confusing?

801
00:56:13,596 --> 00:56:15,716
To ask it to be an editor, in other words.

802
00:56:16,536 --> 00:56:18,656
And in this case, I thought,
well, let me just try voice.

803
00:56:18,996 --> 00:56:22,696
So I enumerated the four items, but
I rambled a bit and I meandered and I

804
00:56:22,716 --> 00:56:24,716
went over here and there and everywhere, and—

805
00:56:26,096 --> 00:56:28,416
It was able to pull out another,

806
00:56:29,496 --> 00:56:33,716
more compelling point out of my
ramble than I had articulated.

807
00:56:34,456 --> 00:56:38,656
And it was more compelling than
anything else, and it won the argument.

808
00:56:39,456 --> 00:56:45,236
And I was just surprised that, wow, I
hadn't even realized how important that was.

809
00:56:45,576 --> 00:56:50,896
It was in my thoughts, but I hadn't
elevated it to be like the crux of

810
00:56:50,936 --> 00:56:52,836
the matter. Does that make sense? So

811
00:56:53,836 --> 00:56:56,336
anyway, I use voice now, and I'm so much faster.

812
00:56:56,450 --> 00:57:00,850
[KAREN] That's really interesting
to listen to, and as you were

813
00:57:01,810 --> 00:57:02,670
telling me about that,

814
00:57:03,190 --> 00:57:08,130
I was thinking, but that wouldn't work
for me, I don't think, because I work so

815
00:57:08,170 --> 00:57:09,530
much better in writing.

816
00:57:11,070 --> 00:57:13,510
And for me, the act of writing something,

817
00:57:14,530 --> 00:57:17,590
and changing it, and rephrasing, and

818
00:57:19,010 --> 00:57:23,850
cutting a bit out and putting something in,
that's how I kind of process my thoughts.

819
00:57:23,950 --> 00:57:24,390
For me,

820
00:57:26,050 --> 00:57:29,850
doing it by voice just wouldn't,
I don't think would work for me.

821
00:57:30,596 --> 00:57:30,696
[CLAIRE] I

822
00:57:31,556 --> 00:57:33,056
would have said that to you a month ago.

823
00:57:33,896 --> 00:57:36,536
I've always been a firm believer
that writing is thinking.

824
00:57:37,376 --> 00:57:39,976
I took a year off to write a
book at one point in my career.

825
00:57:40,736 --> 00:57:44,736
I think of myself as a writer.
I still do my own writing.

826
00:57:44,856 --> 00:57:50,996
I don't outsource writing to any LLM. I
strongly believe in authentic voice and,

827
00:57:52,416 --> 00:57:54,856
you know, that human element, but

828
00:57:56,276 --> 00:58:01,436
I am embracing voice as a way to get
certain types of work done faster, and

829
00:58:01,536 --> 00:58:03,016
I've been blown away at

830
00:58:04,376 --> 00:58:05,856
how much more efficient I am, so,

831
00:58:06,916 --> 00:58:07,256
I don't know.

832
00:58:08,036 --> 00:58:09,576
Just planting the seed.

833
00:58:10,370 --> 00:58:12,130
[KAREN] Okay. I will give
it a try and let you know.

834
00:58:13,456 --> 00:58:13,856
[CLAIRE] All right.

835
00:58:16,036 --> 00:58:21,596
I want to make sure that we have covered
the promised topic of this episode, which

836
00:58:21,656 --> 00:58:25,096
is from Oracle DBA to Postgres contributor.

837
00:58:28,336 --> 00:58:33,036
Actually, in that FOSDEM 2020,
when we were together in Brussels,

838
00:58:33,596 --> 00:58:38,636
I think the talk that I gave there was on
different ways that people can contribute

839
00:58:39,016 --> 00:58:41,936
to the Postgres community. It
was like Fibonacci spirals and

840
00:58:43,056 --> 00:58:45,796
21 ways to contribute to Postgres beyond code.

841
00:58:46,356 --> 00:58:51,396
Because the project, of course, has
documentation and translations and meetups and

842
00:58:52,576 --> 00:58:58,916
podcasts and blog posts and conference organization.
And I mean, there's so many ways to contribute.

843
00:58:59,296 --> 00:59:05,696
Diversity committees, core team,
protecting the trademark, right?

844
00:59:05,796 --> 00:59:08,316
There's a lot of different things there. Is there

845
00:59:09,356 --> 00:59:14,796
anything that you think we should talk about
becoming a Postgres contributor that would be useful

846
00:59:14,836 --> 00:59:15,996
for someone who's listening,

847
00:59:17,216 --> 00:59:18,216
from your experience?

848
00:59:20,950 --> 00:59:21,050
[KAREN] Yeah.

849
00:59:22,650 --> 00:59:26,069
As I was saying, I was very lucky that I was

850
00:59:26,550 --> 00:59:28,410
taken under the wing of somebody. So

851
00:59:28,910 --> 00:59:32,630
Lætitia really took me under her wing
at that first PGConf.EU that I went to,

852
00:59:34,070 --> 00:59:34,830
and I think that—

853
00:59:35,690 --> 00:59:40,050
That's really helpful to have someone to
kind of say, you know, oh, maybe you should

854
00:59:40,090 --> 00:59:41,330
talk to this person about this

855
00:59:44,130 --> 00:59:46,850
and someone just to kind of
point out who's who a little bit.

856
00:59:47,210 --> 00:59:50,590
And I know that there have been talks
recently. I know you've done talks.

857
00:59:50,790 --> 00:59:55,250
I know Valeria has done talks
about, I think she did the—

858
00:59:57,030 --> 01:00:00,210
Is it the Postgres community in a
nutshell? I can't remember the exact

859
01:00:01,170 --> 01:00:01,930
name of her talk.

860
01:00:03,710 --> 01:00:08,670
So there have definitely been a lot of
people trying to make it clearer who does

861
01:00:08,730 --> 01:00:13,110
what and how people can get involved.
And then I'm really, really excited.

862
01:00:13,150 --> 01:00:16,490
I think I've used the word excited a lot
today because I am really excited about

863
01:00:16,510 --> 01:00:20,890
lots of the things going on at
PGConf.EU. We've got a buddy program

864
01:00:21,890 --> 01:00:22,810
at PGConf.EU.

865
01:00:22,536 --> 01:00:28,956
[CLAIRE] Oh, so people can sign up to be partnered
with someone who's more experienced at the

866
01:00:28,956 --> 01:00:30,956
conference?

867
01:00:29,530 --> 01:00:31,290
[KAREN] Yes, and I think we've got

868
01:00:33,220 --> 01:00:35,220
thirty-five

869
01:00:36,050 --> 01:00:40,010
pairings or something like that. I mean, I
might be giving a completely wrong number,

870
01:00:41,630 --> 01:00:46,170
but it was really, really nice to see
how many people wanted to sign up as

871
01:00:46,210 --> 01:00:51,850
a buddy, so to share their knowledge of
the Postgres community with someone and,

872
01:00:53,550 --> 01:00:58,010
you know, just be a friendly face at the
conference, because I've been to conferences years

873
01:00:58,050 --> 01:01:02,070
ago where I didn't know anyone and I
was just kind of looking around the room

874
01:01:02,110 --> 01:01:05,410
wondering if I dared approach anybody.

875
01:01:08,330 --> 01:01:10,750
And I'm sure I'm not the only
person that's felt like that.

876
01:01:12,366 --> 01:01:16,826
So yeah, the fact that we've been able
to pair people up, so someone that's new

877
01:01:16,946 --> 01:01:22,266
or just shy or traveling on their own to
the conference or just doesn't yet know

878
01:01:22,326 --> 01:01:23,446
anyone in the community,

879
01:01:23,946 --> 01:01:27,966
they'll have somebody that they've been paired
up with, who they can meet up with, they

880
01:01:28,006 --> 01:01:31,226
could have a coffee, they could, each pairing

881
01:01:31,926 --> 01:01:34,246
will kind of decide what works best for them,

882
01:01:35,546 --> 01:01:39,926
but yeah, we're trying to put things like
that in place so that people actually have

883
01:01:39,986 --> 01:01:41,706
those little ways into the community.

884
01:01:43,886 --> 01:01:48,066
[CLAIRE] I'll try to dig up a link to that
talk you mentioned that Valeria Kaplan gave,

885
01:01:48,546 --> 01:01:50,986
talking about ways to contribute to Postgres.

886
01:01:52,726 --> 01:01:56,846
And I know that this year there's also
another, I think it's the second year of

887
01:01:56,886 --> 01:01:59,666
this experiment where there's
going to be a community day.

888
01:02:00,106 --> 01:02:05,746
So after the three-day EU conference ends on
Friday, there is the PGConf.EU community day, and

889
01:02:05,766 --> 01:02:11,886
there's all sorts of longer, deeper dive
sessions there, right? So I need to register.

890
01:02:11,966 --> 01:02:14,966
I haven't registered yet. Oh gosh,
I hope it's not sold out yet.

891
01:02:14,966 --> 01:02:18,839
[KAREN] Do it. I think there are a
few places left, so I would do it too.

892
01:02:19,386 --> 01:02:21,246
[CLAIRE] Okay, I better get with the program then.

893
01:02:21,746 --> 01:02:25,866
[KAREN] But yeah, that should be a really
good day. It was popular last year.

894
01:02:25,926 --> 01:02:28,726
We've swapped it, so last
year it was on the first day.

895
01:02:29,366 --> 01:02:31,046
This year it's on the last day,

896
01:02:31,726 --> 01:02:34,806
and I know there were good reasons
for that, and I can't for the life of

897
01:02:34,846 --> 01:02:39,466
me remember what those reasons were.
But anyway, that's how it is this year.

898
01:02:40,426 --> 01:02:43,846
I think it's good to try both and see
which works best for people in any

899
01:02:43,906 --> 01:02:47,606
case. And based on feedback, we can
decide which one we do next year.

900
01:02:48,746 --> 01:02:53,946
But it's, we had quite a lot of people
wanting to organize events for that.

901
01:02:54,086 --> 01:02:56,186
So each one is a half-day event.

902
01:02:57,746 --> 01:03:00,246
So I'm trying to remember, there
are all sorts of different things.

903
01:03:00,286 --> 01:03:04,366
So there are working groups on Postgres education.

904
01:03:06,106 --> 01:03:09,846
There's stuff about Kubernetes. There is

905
01:03:12,246 --> 01:03:19,326
talk about an extensions ecosystem,
observability, automation.

906
01:03:19,365 --> 01:03:21,766
There's all sorts. It's, yeah, it's great.

907
01:03:23,566 --> 01:03:27,346
[CLAIRE] Now, are we going to have
a Postgres 19 release candidate 1?

908
01:03:28,306 --> 01:03:32,246
Yes, we are. It looks like
it's due out later this week.

909
01:03:33,066 --> 01:03:38,566
Obviously, GA isn't going to happen until
November this year, mid-November-ish, I think.

910
01:03:38,646 --> 01:03:42,886
I'm looking at the latest dates. But release
candidate 1 is targeting later this week.

911
01:03:43,706 --> 01:03:44,866
So that's very exciting.

912
01:03:45,586 --> 01:03:46,526
[KAREN] That is exciting.

913
01:03:46,806 --> 01:03:52,846
[CLAIRE] I think people who have worked extremely
hard on the annual release will be happy

914
01:03:53,686 --> 01:03:55,226
to be so close to the finish line.

915
01:03:56,286 --> 01:03:59,406
I don't know if they'll be in a celebratory
mood yet because it won't be over

916
01:03:59,466 --> 01:04:02,226
yet when we see each other next week, but...

917
01:04:04,346 --> 01:04:11,206
[KAREN] And I suspect people that have written talks
for PGConf.EU about Postgres 19 features may be

918
01:04:11,246 --> 01:04:15,046
feeling a little bit calmer about
their talks if we're at that stage.

919
01:04:14,786 --> 01:04:18,806
[CLAIRE] Oh, things are more settled in terms
of the final contents of the release and all

920
01:04:18,866 --> 01:04:24,146
of that. So one of the cool things
that happened to me in New York City,

921
01:04:19,846 --> 01:04:20,386
[KAREN] Yeah. Exactly.

922
01:04:24,626 --> 01:04:28,906
[CLAIRE] the annual PGConf New York City event,
which was renamed this year, so it's now called

923
01:04:28,966 --> 01:04:33,346
Postgres Summit US, just happened, and Joe Conway,

924
01:04:34,106 --> 01:04:39,386
who serves on the contributions
committee for Postgres,

925
01:04:40,626 --> 01:04:42,626
they've started giving out coins.

926
01:04:42,666 --> 01:04:46,186
There's always been coins for
anybody who contributes to a release,

927
01:04:47,006 --> 01:04:49,046
and those are called challenge
coins, and it's pretty cool.

928
01:04:49,086 --> 01:04:53,586
And folks are—there's a new one for each new
release of Postgres, and people are always

929
01:04:53,626 --> 01:04:57,406
very excited about that. But
there is also now a one-time

930
01:04:58,106 --> 01:05:04,006
significant contributor coin and a
one-time major contributor coin. And so

931
01:05:04,246 --> 01:05:04,366
[KAREN] No.

932
01:05:04,846 --> 01:05:10,646
[CLAIRE] I received my significant contributor
coin just two weeks ago in person, and

933
01:05:11,246 --> 01:05:13,126
I'm sure you haven't gotten yours yet, have you?

934
01:05:14,666 --> 01:05:19,486
Okay, but you are a significant contributor,
and I'm sure that Mark Wong will have hopefully

935
01:05:19,826 --> 01:05:24,786
a coin for you when you get there. So
anyway, it's just, it's kind of cool.

936
01:05:25,726 --> 01:05:26,247
[KAREN] That is cool. Yes.

937
01:05:26,446 --> 01:05:33,766
[CLAIRE] There's something really special about
recognition of work on an open source project. So,

938
01:05:34,346 --> 01:05:40,486
and since this episode is titled From Oracle DBA
to Postgres Contributor, I thought it fitting

939
01:05:40,906 --> 01:05:44,586
to end with a thank you to you for
all the work you've done for the

940
01:05:44,626 --> 01:05:51,266
Postgres project, to give you that physical
thing to look forward to. And here, we're not

941
01:05:52,206 --> 01:05:56,786
on video on the recorded episode,
but just so that those of you know.

942
01:05:57,806 --> 01:06:02,306
Just so that those of you know how
the sausage is made, we are on video

943
01:06:02,446 --> 01:06:06,866
backstage. It's just the video isn't
recorded, which is good because

944
01:06:08,606 --> 01:06:10,726
I'm not exactly having a good hair day here.

945
01:06:11,446 --> 01:06:15,686
But I'm trying to dig up and show
Karen the coin, and I guess I'll have

946
01:06:15,726 --> 01:06:18,086
to do it later. It's too hard to find it now.

947
01:06:19,466 --> 01:06:23,986
But I do thank you for everything you've
done to contribute and are still doing. So

948
01:06:24,506 --> 01:06:26,086
the journey is not over yet.

949
01:06:26,746 --> 01:06:28,086
[KAREN] It's not, no, it's a pleasure.

950
01:06:30,086 --> 01:06:35,046
[CLAIRE] And thank you for joining us today on
this episode. I really, really appreciate it.

951
01:06:35,806 --> 01:06:36,606
Glad you're here.

952
01:06:35,946 --> 01:06:39,626
[KAREN] Thank you for inviting
me. It's a fitting way

953
01:06:40,926 --> 01:06:41,706
to mark the

954
01:06:43,046 --> 01:06:43,446
kind of

955
01:06:44,406 --> 01:06:47,906
coming back to normal of my family life after

956
01:06:50,446 --> 01:06:52,586
the attempted recording last time.

957
01:06:54,186 --> 01:06:56,646
[CLAIRE] Well, thank you to
Karen Jex for joining us. And

958
01:06:57,126 --> 01:06:59,546
if you liked today's episode
and you want to hear more

959
01:07:00,246 --> 01:07:06,006
of these Talking Postgres shows, you should
subscribe on Apple, Spotify, YouTube, or wherever you

960
01:07:06,046 --> 01:07:09,086
get your podcasts. And
please tell your friends too.

961
01:07:10,126 --> 01:07:15,226
Word of mouth is gold in the
podcast world, and people generally,

962
01:07:15,686 --> 01:07:18,706
if they like the show, they end up
being really appreciative of it.

963
01:07:19,186 --> 01:07:24,046
You can always get past episodes and get links
to subscribe on the different platforms at

964
01:07:24,506 --> 01:07:30,886
TalkingPostgres.com. And transcripts are included
on the episode pages on TalkingPostgres.com too.

965
01:07:31,506 --> 01:07:36,006
And a big thank you to everybody who joined
the live recording and participated in the

966
01:07:36,046 --> 01:07:37,606
live text chat on Discord.