Subscribe
Share
Share
Embed
Welcome to AutomationTown! A podcast about regular people, building automations for the problems we all share.
First time here? Start with S02E01 "For When You Need A Chatbot"
What are the things you do each day that you'd rather forget?
Chad & Jason explore common pain points for knowledge workers, and track down the people & tools necessary to automate trivial to-do's.
Previously on Automation Town with 29 of 31 precincts reporting, we are officially able to call it Chad Davis will become the next mayor of Automation Town. Mayor Davis and I were just discussing the new Joint Water Initiative Water Initiative tomorrow. You don't think that's
doors? The control room in the subway tunnels. Uh, Jake created the water disaster. Why would he give us the solution? Goodbye for now, mayor Davis. I see bright things in our future.
I remember the days. Good morning. Good morning. So are you ready to save the world again? Again? Another day. Another door to kick down. . You guys are in trouble.
Yes. My batteries are dead. Okay.
Orange juice, grits and bacon. Thank you dear. Yes, mom,
Phil. Meet Gill, the new director of finance. Phil Woo. And I'm the director of finance. I haven't been able to pinch Chad down on the details. Gil, you're over here. Well, I, uh, look forward to working with you. What.
That was all I said. You didn't talk about the water pipeline with him at all? I mean, I wanted to, but we didn't have time. So when he announced the initiative during the speech, I was as surprised as everyone else. That is wild. Wow. Hey, hey. The clocks? Mm-hmm.
Look at that. The whole band together. Well, it's happening again. What's Jake MCC Kringle Berry up to now with this water pipeline that connects the two towns. This feels an awful lot, like a, like a feelgood montage before the whole finale sequence kicks off. It sure does. I was about to say the same thing.
And data processing. Data processing. What do you do when you gotta crunch some NUEs or take some data that's in one format and convert it to a different format? We can automate that. Oh, we can automate that. Let's crunch some. On this week's automation town.
Oh, oh yeah. That's just rv is this thing Parked on a sinkhole is not what she wants was I've noticed that when we have a bunch of people in the rv, it's making some, some new noises. I told you Wesh brought McClue. It's the McClue. It's not the mcle. Thank you, by the way. Okay, so on the subject of Saving the World, the new Water pipeline from Manuel to Automation Town goes live tonight.
Sometime tonight, I still need to get the full briefing from Cat, though Amelia and I can head down to the tunnels to check it out. I'm thinking we head back down through the Radio Shack, then through the control room, see if we can get any sense of what's going on, and then go. I feel like we should be leveraging some sort of mayoral access or surveillance supercomputer to get us answers.
Now, Davis, I'm not sure that's a thing, to be honest, I've hardly had the time to be. Mayor Kat's been on me to connect though, and after this recording I'm gonna meet with her and hopefully get a better handle of what the day-to-day looks like. Yeah, any supercomputers? Yeah, with the super computers, skeleton keys, skeleton keys.
Okay, so the show today, data crunching analysis, reformatting all things data related. Oops. All data. Hey Pat. Hi Chad. I've created a view for all the questions in the database related to processing data. Would you like me to start recording? Oh wow. Hey, guess they don't need you anymore. I don't know what to tell you.
The other night we were talking who's we? Pat and I, I swear I've seen this movie before. He offered to integrate our podcast recording software. Sweet. It's okay, sweetie. Is. What it feels like to be automated. Paul, you are doing more than hitting the record button. Okay. Everybody in position? Hot mics. Hey, hot mics, friends.
Welcome into another installment of Automation Pod. I'm here with my friend, co-host, major local celebrity has Excellency, the mayor, Chad Davis. Good morning, Chad. Good morning, Jason. Today we're looking at data processing. We're pulling a bunch of questions about data processing so that the next time you need to process data, you've got it down.
Pat. Pat here. Wow. Who's ready to process some data? I like the enthusiasm. A question from Georgie. I'm using API connectors like Zapier and make to move data from A to B, but I find when I have to change that data or formatted a certain way, that can be frustrating to figure out. , do you have any best practices for me?
Uh, formatting. I feel like formatting's, that one topic that's like the end of an automation. You feel so good that you've connected these thing together and then they come into your spreadsheet or your database and it's like a timestamp with a code that's like 75 characters. Long . Yeah. And you're like, uh, what do I do with this?
And then you Google it and you're like, Oh, I actually have to make it look how I wanna make it look. Zapier and make, have their own built-in stuff. Uh, talk to us about the Zapier side. Uh, how would you kind of clean things up if you needed to, if you were just starting out? Yeah, my frameworks kind of start with the most simple solution and then work downhill from there.
Uh, what Zapier will do, I think what you usually end up using, It's an action step called Formatter by Zapier. Basically, it's just an action step. Just like any other action step in Zapier, you tell it what type of data you're trying to format. So the options here, let's take a look. You've got date and time numbers, text, and then utilities, which is like a catch-all for everything else.
But to build on your example date and. . It gives you a bunch of different functions you can use on dates and time. So add or subtract time, compare dates, formatted a certain way. Let's do format, cause that's what you're talking about. You give it the input value from presumably the previous step, and then you just say, I wanna structure my date and my time this way.
I use date time quite a bit on zary. The other one I use a ton is, Just number formatting actually put in some decimals and some thousands and a currency signal on it if you need to. Yeah, and like the commas and like you just, you don't realize there's actually a hundred different ways for a number to be formatted and inevitably it's not the one you want.
But Zappier works pretty well for like single numbers. I was actually working on something today where we needed to. Bring a number into an automation and then do something with it somewhere else, and it was in like this perfect number format. So I thought, and then it comes in and you run your automation and then you get an error.
You're like, that number's not a number. So there's more to formatting than, you know, changing the thousands. Like formatting can mean, you know, put it in text, put it in a number, put it in the right, you know, format for actually working through the automation. in any one of these things like learning the difference between strings, which typically text and numbers can make a big difference in whether or not a automation works or doesn't work.
And it only takes a few minutes to figure out if something doesn't work. But that was a pretty good unlock, trying to make sure that, you know, numbers are numbers and text to strings, and you kind of format it into however you need to for the automation to actually. . So we talked about zappier. What's the equivalent on the ME side for formatting?
Yeah, so make has some actions sort of built right in when you are entering fields into different modules. So say you're sending an email out to somebody and you wanna make sure that there's no spaces at the end of an email address where somebody might have put a space on your form from a website or something.
A lot of these actions you might not have like a specific step. like Zapier does to reformat it, but you can reformat it however you want in any module. So for example, in that one, you might put a trim action on top of that one field and it'll remove all the spaces. But if you're gonna reuse it over and over and over again.
Mm-hmm. , maybe you set the beginning sort of variable to the trimmed version, and then you can just use that all the way. They have a lot of other ones called, uh, like for this number example I was talking about before, we used something called, uh, parse number. And that will take anything that's a string or something else and turn it into a number.
So once you get your, your feel for like certain different actions inside of make for these types of reformatting, uh, it becomes pretty easy to just quickly change it. to do it once in the beginning and then have it run all the way through. Good example with Zappier is that if you formatted it in one format, That's how it works throughout the rest of the entire Zap.
If you needed that same field in a different format, you'd have to run another format in the middle. Right? A format of the format? The format of the format or, or just the second format. Whereas in make, you can get really granular right in the actual module and just pumble on a whole bunch of different formatting steps right in the actual field for like the email or the amount or the description or something like that.
Good example is like capitalizing things too. One other fall. if you just can't get it there. At the risk of sounding sweaty is code by Zappier. Uh, you've got the ability to just like, if you know a little bit of Python or JavaScript or something like that, sometimes it's just easier to get it there because you know how that works already and you know how to write it already.
There's also some things that are just won't do. Mm-hmm. , getting somebody to write you like a one or two liner, that can be a good shortcut. Chat G p T has honestly made this a lot easier to say. This is this thing that I've got, here's what I want it to be. Write me JavaScript or Python that will convert it to that.
And honestly, most of the time I can like copy, pasta that thing straight in and it works totally fine. You're 10 times smarter after November, 2022, writing code Jason and also simultaneously dumber in other otherwise yeah, so Zappier is great cause it's got it's got code by Zappier. Bill Wright in super handy.
The closest thing I've found to this is a service called One SaaS. The number one s a a s, and that will basically give you these little standalone nodes that can be little blocks of code, almost like a itty bitty Lambda function or, or like just this little thing that's part of your make scenario. Yeah.
That's one of the, the major asks right now in the May community. It's like, Hey, give. A JavaScript module. We just want that, just like Zapier has. Okay. Stop saying JavaScript here. Everybody's gonna stop listening. Uh, pat, what else you got for us? How do the two of you think AI could impact data management in the future?
Oh, he said ai. I like this one. What do you think, pat? You have 114 days. Well, that's ominous. Uh, Chad has, has the advent of like normy accessible AI changed how you're hustling data at all yet? I guess we did just talk about one example, right? Like helping you with little scripts. Yeah. I mean, I am not a script writer, but using.
AI right now. All those little unlocks of Google apps, things that you couldn't do before, and sending web hooks and these little scripts to format things, it just becomes so accessible now. And I think the great thing about the data management aspect of AI is that if you don't understand. Something you can ask it and it'll explain the concepts to you until you get it.
And I find myself constantly in super inquisitive mode being like, okay, I don't understand this. Step back five steps and, and walk me through exactly how this works. And can I just, can I cut you off and like double down on how smart the thing they sang right now is because last episode we're talking about web hook.
And it's just hard to like verbally explain certain things, but that's such a great use case for chat. G P T right now is I'm a du. Tell me what this thing does. The more you go back and forth, the better. Because like you can be like, I don't know what this bit or that bit meant. It's such a great use case for it because it's not gonna make fun of you cuz if you're asking the dumb question, it's not, it's not going to overtly, um, judge you.
But behind the scenes it probably is anyways. What are we talking about? AI for data processing. Have you ever played around with, uh, chat, G P T and just data stuff? I have. I don't know if it's a good idea or not. . It's pretty darn good. It's chat. G b T isn't great at math, even though they're improved it lately.
It's still not amazing at math, but I've done stuff like, here's a CSV with 20 records of data. Apply this transformation to it. Add this column that can catonate, you know, two other columns or something like that. Just semantically does all that stuff fine. Doesn't break a sweat, you know, filter a view.
Here's a named view. I may refer back to it later. Um, just like applying all these transformations to a set of data and it'll present it back to you in a table format that looks like a spreadsheet. And I've done this up to like eight to 10 transformations and then said, okay, here's a new set of data.
Apply all 10 of those transformations. And it usually doesn't just fine, which is pretty wild. I don't trust it enough to do that in a mission critical situation. But I think what's really exciting is like the notion of semantically being able to build robust, durable, like stuff like that because data transformation tools, like start talking about stuff like R and like really technical stuff, like it is so out of reach for normal people.
Yeah. I. I got stuck a few times where, you know, Chad Gpt would write a script and you'd use, you'd be like, uh, I, I actually can't get it to do it exactly what I want. So, but it got me close enough and then I'd throw it into chat G p t and be like, uh, remove that column, or something like that. And it would just do it like, okay, cool.
So like this whole mechanistic scripting world is there, but wow, is it so much better to just. To it to tell you what you want it to do. And I mean, I'll be the first to admit. My spreadsheet formula game is like a hundred Xed in the last three months. Wow. Really? Now you're doing sequel queries and like really complicated filters and like all these things that would just take a while to figure out and get.
Right now you're just like, do this thing for me and it does it for you, and you're like, oh, that actually worked. And when it doesn't work, you tell it, it doesn't work, and then it fixes it. So like , I dunno. Feel smarter. Yeah, spreadsheets are a great example. The other day I needed like, I need like a list of unique instances of things and it did that for me really well.
But I think the holy grail is just semantic, being able to explain what you want and then reuse it and it just works. We're not there. Uh, but like I saw an app in Early Access the other day that is like a chat G P T interface that integrates. Airtable and Postgres and Stripe and a whole bunch of other data sources and even like Zapier and Will like triggers apps for you and like you can say, run this or that automation and push this data from here to there.
Like that's gonna get really wild. It's hard for me to trust it cuz I haven't actually seen how it works yet. But all that is to say, I think AI will make the notion of building your own automations and all that much more accessible. Oh, incoming. Recording paused, Gil. Oh hey. Whoa. This is how it ends. Oh, this poor RV Chad.
Uh, quick question about Phil. Who's Phil? So he's the director of finance. Ooh, yeah. I didn't know we had a director of finance. Yeah. Ugh. Okay. Well I'm meeting with Kat later on, and I guess we'll need to sort that one out. Okay. One other thought on live flow. Of course. Live flow. So it turns out the city has spreadsheets on spreadsheets.
On spreadsheets. Have you seen these things? I haven't. From what I've gathered, they used to do all the fiscal stuff on these massive ledger sheets, and only recently transitioned them to a spreadsheet. That's just a digital version of the same thing. Okay. And it got me. I could use live flow to populate the spreadsheets they've already built with live data from our QuickBooks file.
Yeah. They've already got this reporting set up. No use reinventing the wheel for now. You could use live flow to eliminate manual entry on those reports. They're already preparing. It looks like each department has their own master budget, but then I've talked with several staff that are each responsible for their own sub reporting.
Sounds like a lot of spreadsheet updating. Yeah, and it got me thinking not just for the city, but for my coaching clients. Live flow's a great solution for automating reporting when you've already got reporting. Yeah, I suppose it doesn't involve moving to another platform. If you've already got reporting you love or that you've built some infrastructure around.
Live flow just makes it smarter and it'll keep syncing live data in Right now, they update those reports monthly, but there's no reason that we couldn't now update them daily since live data sinks in from live flow and it doesn't require any human entry. I love it, Gil. Good stuff. That's why I made you my director of.
Yeah. So did you though, uh, am I the director of finance? Hmm. Because Phil, yeah. Let me connect actually. Speaking of Kat Cat, um, mayor Davis. Oh God. Not like this. What is this? Okay, now we might be over capacity. Hey Kat. Meet the crew. Crew. Meet Kat, chief of staff. She's of Hi Cat. Hello. Hello Cat. Hello crew.
Gail, what are you doing here? Well, I was, Chad and I were just talking about Phil and I told you Mayor Davis and I would discuss this later. Of course, of course.
I'll be going then. My See you later, man. Love you. Hey, pleasure as always, gang.
Oh, that makes it down. Woo. So we're recording the podcast right now here? Yeah. I guess I imagine, Hmm. It's cozy. What can I, uh, why are you here, Kat? Mayor Davis, you've been in office for three days, and in that time you've only been to said office once for about an hour. So I'm here to corner you. Let's meet here.
I'll admit the marble halls of the capitol. I gotta take some getting used to. You're welcome to make yourself comfortable. We won't be too much longer here, Kat, take our spot. If we're gonna save the world tonight, Amelia and I need to get ready to go down to the tunnel. Pat, once you're done recording, if you could please create a folder for this episode using the correct episode ID in the drive.
Then save the files to that folder. That'd be great. Rajak. Wow. Paul's putting himself outta work now. Mm, it's looking that way. See you guys later. Pleasure to meet you. Ka.
Did that gentleman say save the. Uh, yeah, let's put a pin in that and let's wrap up the recording here. We've got a couple questions left. Pat, where were we? You just answered a question about AI's impact on data processing. Right. Let's start recording and. Please read out the next question. This question is from Lisa.
Hi guys. I have a recurring workflow where I need to process several hundred lines of data a certain way. We've historically just done this in spreadsheets, but it ends up creating a bottleneck when we need the processed version on a timely basis. Any ideas for processing Thousands of lines of data.
Okay. Hundred lines of data. This is a little different than. Putting a comment at the end of a sentence or capitalizing your mom's name. Yeah, . And plus these systems cost money, so you have to take into consideration price when you wanna run a whole bunch of stuff through it. Are we in agreement? You wouldn't run something like this through Macon Zap here?
I don't know. It's funny cuz there is a point in time where, Maybe a couple hundred is worth putting it through, something like that. But maybe a couple thousand isn't, just maybe depends on how heavy the lift is. So you know, you need to update a hundred invoices or something. Yeah, you're probably gonna throw it through make, but if you need to do some transformations where you're just formatting something, probably not, you're probably gonna try to find a different way to do that before you start your automations or put it into other systems.
Is that how you think too? Yeah, I guess it's worth delineating between. The, like the use cases where I think spreadsheets are still fine and when they're not, and if it's a recurring workflow where you gotta do the same thing on a recurring basis, like yes, look for a way to get it off the spreadsheet.
If it's like a one-off thing, like ad hoc, like who cares? Like it's not worth setting up the automation for. Yeah. But I do know there's a lot of those like ongoing needs where you've gotta take data outta system, egg it in a system B. And oftentimes the fact that you have to do something manual keeps that from happening in the timeframe where you want to.
There is some like spreadsheet manipulation that you can do with Macon, Zapier. It's pretty rudimentary. My next step is to look at. Is it something that I can do? An Airtable, like my kind of mental framework is Megan Zappier usually gets me by with like a small volume of data. Airtable gets you by with a pretty darn good like bit of data.
Uh, and then the built-in Airtable automations are, are super powerful. If you can add that with. Air table scripting. Occasionally you'll need that. Um, but usually you can get by modifying your data in whatever format you need, just with their built-in automations. Another sort of wrinkle here is, is oftentimes what is the output format that you want it in?
And if historically you've done it in a spreadsheet, the expectation of the people on the receiving end oftentimes is, oh, I want this spreadsheet, this formatted and highlighted just this way. If it's beyond what Airtable will do for me in terms of record volume, then I'll usually look at like SQL and Chuck It and Retool or something like that.
Retool's a helpful little interface for working with SQL Files, but there is a like similar to Zapier dedicated app for data processing called parabola. That I think is probably the leader in this space right now for automated workflows, just for your data. It's funny when you, when you look through that stuff, you're like, yeah, these are, these are also like brand new skillsets, like SQL and retool, like they're hard.
Some, some of these things are more, they're expensive as well. So if you think about like the accessible options, this, you know, Lisa has a recurring workflow that happens all the time. She needs to get it into a certain format. I think where my head goes to some of these things is like, what is the simple lift?
And like two options come to mind. The first one is, Most people have a Microsoft Office, uh, subscription or, or, or plan, right? So inside of Excel, you know, there are functions that are built for this, right? You drop in your CSV or your Excel file into a folder. Yeah. And then Power Query will add that data.
To the data set inside of your master sheet, like that is a huge lift. It's still very spreadsheety and not very automated, but it will probably get you to a point where you have everything in a place that you need it. But what I find myself doing more now is making it accessible for the people that have to do the work and instead of getting them to download and to manipulated and upload it and do all the things we like, we normally would do, give them a.
give them a, a simple web form where they can drop it down, maybe pick a date range, have their name or something like that. And then like, that's all they have to do. And like there's a really, you, you mentioned this at the very beginning of the question, you said, you know, there are some li lifts that Zapier and make have mm-hmm.
one of my favorites is inside of make, you can take the form submission data, which can be all in CSV or in whatever it is, and then create a new Google drive. From that data. So instead of going through 10,000 lines or 20,000 lines, it just. , boom. One operation, right? It creates the file and now you have something online that you can go through on an automation and actually like use if you needed to.
So I find myself using that a lot. I think this is a tricky one to answer because people use spreadsheets in so many different ways, and the right answer is very contextual on. Is this something that has to get processed immediately, like on a fully automated basis? Uh, is it something that you're doing a bunch of times?
Is it something that happens once a year where it's like, who cares? Just hop in, do it in a spreadsheet, especially time sensitive things. There definitely is value in like wiring up a fully automated end-to-end thing, and that's where para is gonna be suited to do that job for you because it'll take it from a.
As soon as that data's ready, do the transformation and put it in B where it needs to go without anybody having to touch it. It is one of those things where it's like another tool to learn though, and it has a little bit higher of a bar to learn than something like Zapier. Uh, Chad, do you know what this feels like a segue.
Let's talk about our podcast sponsor. Cinder Cinder. Oh, cinder. So processing thousands of lines of data. This, this is just what Cinder does with e-commerce data. Oh, e-commerce data. Yeah. So eBay, Amazon's Stripe, huge list of integrations, but getting e-commerce data back to your accounting system, it's a big oof a huge project.
There's so much data and there aren't generally helpful native integrations to get that data back to. And that is the beauty of Cinder. It not only syncs that data over for you, but it gives you control over how that data is synced over. Yeah. Weren't you saying you can do it either by, what was it, transaction or batches?
Yep. So let's say you're running itty bitty Etsy shop. You've got five to 10 sales per week running through like nothing huge. So in this case, I'm probably gonna have cinder sync every single transaction over to the accounting file. Okay. But if you had something higher, a lot higher volume, let's say you got a Shopify store running a hundred sales a week thousand sales, a.
Instead, you'll probably wanna pull in just a daily summary, and that's the only thing that goes to the accounting file, because pulling every single transaction would be a mess. In that case, yes, but the main upside here, you get to choose. You can tailor each of these integrations in the way that makes sense for your data.
Very nice. Give the people the website, Jason, the website, cinder, s y n d e r.com. Check 'em out for wrangling your e-commerce data.
One more based on the recording Time elapsed. You have time for eight tenths of one question. Okay. Weird. Pat, just give us, give us another question from Matt. Any suggestions for data exploration? I have a need that's a mix of data processing and data exploration. I still need to be able to see and manipulate the data.
Okay. We're explorers now. Davis, how do you explore your data? Uh, it's a loaded question because it depends on how people use it, right? If it's just for yourself, you can typically wrangle around and make weird little formulas, and if you don't have to explain it to somebody, you can just kind of make the worst looking thing ever.
But as soon as you have like two or more people that are gonna look at this, you have to start thinking about how clean this is. How much can you trust the data? And. Like you have to think, okay, is spreadsheets the actual best way to do this? And for decades for the small business out there, like that's the answer.
But now there's other tools out there and with a little help from maybe. an expert or uh, a contractor to come in and help you. You can really leverage a lot of this tech out there and have your own way to explore the data in whatever form you want. Chatbot, uh, dashboard, email digests, uh, a spreadsheet that's linked to all this kind of stuff, like the world's your oyster, I guess with the right kind of person building it up for you.
Um, I know that doesn't answer your question, but there's. I think it actually is just go towards your audience. Whatever your audience is looking for and needs. That's kind of where you go. And it's, again, not all spreadsheets. So think about it through the lens of the destination, like what's the, what's the end user gonna need, because that is true.
Like there's so many times where, , you're like, I just made the most sweet thing you've ever seen , and everybody's gonna be so impressed by this, and I just spent a whole day on it. And then you give it to somebody and they're like, what the hell even is this? I don't understand what it is. So, or they just want one number.
Yeah. Yeah. No, yeah. No, I didn't actually need any of this. That definitely is worth keeping in mind. Like what do you actually need on the other side of it? That's another question where spreadsheets mean so many things. That could be 20 records of data. It could be 20,000 records of. And so the right answer definitely depends on volume.
Google Sheets now will technically handle like, I think, up to 2 million records of data. Excel, well think starts chugging a long ways before you get there, but when it comes to like exploration, there's definitely a bit more of a technical bar there. I will say, we talked about Simple ML the other day. It was a extension for Google Sheets that will do some machine learning types of things to Like predictions.
Like predictions? Yeah, like predictive stuff, finding anomalies. Mm. That's one thing where exploration for a human is gonna look different than like pulling in more AI enabled tools because that's not a thing that you could really do yourself. I have to imagine there are tools to assist with exploration for certain types of data, and if there isn't, maybe you look at something off the shelf that's generic, like a simple ml, but.
I would say these days, if you have a set of data that's any more than, I don't know, 30 records, look and see if there's an ML tool out there that will help you, like detect like, what are the outliers here? Just give you a more meaningful way to dig through it than just filtering and scanning through.
But then you think about the work that you do on a, on a daily basis, like, uh, somebody just wanted to know. , you know, which bills to pay or, or if we have to hire or not, or who do I have to reach out to for the dog walking business? And like a lot of this is, is simple. You probably wouldn't use these, these ML tools for, which are great, but I think that like how do you communicate this to other people and like, Even, even if it is spreadsheet or like little quick front ends to databases that are spreadsheets, like that's the real win I think over the next kind of couple years is giving people visibility without having to teach them what you did during that full day to make the spreadsheet work.
and like, that's what gets me excited about all this stuff is that, that now it's, it's so accessible with chat G P T or with AI or with, you know, any conversational AI tool to learn how to do this that you might not even need to know about the app. You just ask it, what are ways that I can share this information?
And then they give you these things and it's just, It's a huge unlock, so I won't be trying our anytime soon. , oh man. I dabbled a bit in college and that's just a level that my brain's not operating on. Okay, I think that is all we've got time for today. Thanks so much for tuning in. Keep automating and stay safe.
That's a wrap. Nice job everyone. Wow. Just like Paul. Okay. I can get outta your hair. Kat. Pleasure to meet you, Chad. Let's regroup once you're free regarding world saving. Yeah. I'll let you know.
Oh boy. Where to begin? I have an agenda here of what we need to cover. Wow. I told you we've got a lot of ground to cover. Okay. This all looks great, but let me ask you, this whole water initiative, Yes. Tell me how that came to be. Well, when Mayor Mccr came to office and started digging
flashlights check and warmers check, trail mix with the M mss check. Whoa, what was that? Oh, that's just, I've got some sardines. Ugh. Quick and easy meal. Lean protein, Paul. Okay. The plan, gain access to the subway tunnels. Yep. Find the control center we saw last time we were down there. Then what we should see if anything's changed.
Maybe we will see if this has been controlling the pipeline from Berg or not, because pipeline from Manuel Berg bad. Jake created the automation town water disaster, so it's just so weird that he would create a solution for it unless he's up to no good. Yeah. Ready to go. Let's save the world.
Welcome radio. Oh, hey Paul. What's all this? This picture? Oh, buzz. Who is? That's my manager. Your manager is Buzz Mc Tompkins? Yeah. He owns Radio Shack. He isn't here too much. Buzz owns the Radio Shack.
Hey you. You can't go back there. Yes, we can see you later. Hey, that's for employees. Okay. Down the hatch. Nice. Down the hatch.
It's this way. What do you think is that way? I don't know, but that's a question for another day. Look at us. We're just a couple of us. Super cool. Super Spice . Yeah, something like that. I always thought the spy life would be more like punching and business class. And business class. Yeah. You know, like private flights, first class wheels up in an hour sort of thing.
Wheels up in an hour. Martinis on the airplane. Oh, we are audio.
Automation Town is written and produced by Chad Davis and Jason Staats edited by Paul O’Mara. Keep up with the characters of Automation Town on Twitter @AutomationTown.