Subscribe
Share
Share
Embed
If you're freelancing as a software engineer, you've probably noticed something a bit weird: your freelance pricing is now working against you. The faster you go, the less you bill.
This episode is a conversation with Judith Bohlert — a freelance software engineer based in Germany, self-employed for 3 years, building greenfield MVPs and prototypes for early-stage startups — about how freelancing with AI tools affects the job, pricing, and the identity of an independent engineer.
We get into the case for moving from time-seller to product engineer-consultant, why a productized service or a value-based offer is now the obvious response to AI freelancing economics, and why your personal brand — not your stack — is the only moat left when average code becomes a commodity.
TIMESTAMP
0:00 — How AI changed a Monday morning for a freelance dev
4:15 — Tools that actually work: Claude Code, Cursor, GitHub Copilot
5:41 — Repricing your work: hourly vs fixed vs productized vs value-based
17:46 — Use AI to work less, not more: the case for the four-day week
22:04 — Personal brand as the only moat when code is a commodity
35:00 — The freelancer label is dying, the consultant role is rising
38:54 — Advice if you want to start freelancing now
GUEST
Judith Bohlert is a freelance software engineer based in Germany, with around 3 years of self-employment experience, focused on greenfield MVPs and prototypes for early-stage startups. She writes the low noise newsletter — a calmer take on working in tech, without hustle culture or the productivity-hack treadmill.
Website:https://www.judithboehlert.com/
Newsletter (low noise newsletter):https://www.lownoiseclub.com/
LinkedIn:https://www.linkedin.com/in/jboehlert/
CONNECT WITH PRODUCT ENGINEERS
Host: Peppe Silletti
LinkedIn:https://www.linkedin.com/in/peppesilletti/
Peppe's website:https://peppesilletti.io
Product Engineers Community:https://discord.gg/4sMtRNSgU4
Share this episode with a freelance developer friend who's billing hourly and starting to feel it.
This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit
Product Engineers is the podcast about the convergence of product and engineering in the AI era: what's breaking, what's working, and the craft of doing both well.
Welcome to another episode of the Product Engineers podcast. I am Peppe, your host. [rock music] We build it, we own it, we ship what comes next. Welcome to the Product Engineers podcast. Wow.
If you're freelancing as a software engineer right now, I'm sure AI didn't only change how you work, but also how your business looks like.
You know, the projects that used to take maybe two weeks, now they take three days, but you're still billing for those three days.
Your clients are super happy, but then you walk away from the gig with this nagging feeling that something is a bit off, right? And that's really the problem.
The hourly rate and the daily rate, they are pricing models that make sense before AI, when coding was the bottleneck. But, uh, now that doesn't make much sense anymore. So what are clients actually paying for now?
And what are you supposed to charge for it? To explore this question, I sat down with Judith Bolart. She's been freelancing in software for years.
She has lots of experience helping early-stage startups build their products, and recently started a newsletter to explore how to work in tech without burning out.
By the end of the episode, you'll have a, a sharper picture of what the freelance engineer's job i- is actually becoming, why your CV might matter less than you think, and also one idea about what to do with the time AI gave you back.
I'll see you there. Welcome to the, to the podcast, Judith. This is, uh, nice to have you here. I'm very excited about our conversations today. Thank you. Happy to be here. Yeah.
Yeah, um, I'm very excited about, uh, talking with you about freelancing and how it's, it's changing, uh, s- since AI came, came to, to the world, right? And I know you have a lot of experience with it.
You've been doing it for, for a long, a long, a long time. Um, so, you know, the, the first thing I was wondering is if we think about the pre-AI era, let's, let's take like Monday morning. Mm-hmm.
So, um, could, could you walk me through like what a day looked like back then, a Monday morning compared to now? What, what changed? What are the most different things? Yeah, good question. A Monday morning,
I guess typically there's something like a, a standup with your team where you just align on the work that has to be done, maybe a little bit of back and forth with the, the product or project managers on what exactly has to be included in a certain feature.
And then you just get to work, I guess, and you just, um, you write your code.
You try to be very clear on what you want to develop, but then you actually write every single line of code, and I guess that we don't really do anymore.
And then usually on a typical Monday morning, there would be a lot of time spent debugging because there's always something that doesn't work. There's a bunch of error messages, so we would go into debugging.
Also, one thing I've realized just lately is I, I work a lot with like super early-stage startups. I build often like the first MVP or prototype, so I do a lot of greenfield projects.
And typically when I would start a new project, the first one or two days were kind of spent on like bootstrapping the entire project, setting up things like formatters or linter rules.
I feel like we don't really do that anymore. Like that has become far less important whether we use a semicolon or, um, single quotes or double quotes.
I feel like these, these tiny little things that we kind of used to obsess over a little bit, they've become less important in the AI era, I would say. Okay. Okay. So w- what's, what's your, uh, typical client usually?
Because I feel that also can affect how your work looks like. So you, you were saying early-stage startups where I expect like, uh, smaller teams, so where communication is easy and everything is easy.
Did you also work with the bigger clients or with the projects that are already, you know, out there? Yes, definitely.
So sometimes I also work with bigger clients or more like corporate clients, but then I would often work with them when they are trying to build like a new tool within their company.
So I don't work a lot with like super huge existing code bases. That happens sometimes, but it's just not, not my expertise really. So I really often start from scratch
with greenfield projects, but even that is often quite different whether you work with like a small scrappy startup or a big corporation, um, especially when it comes to AI.
Because working with like larger corporate clients, they often have like very strict rules on how you can use AI, if you can use it at all, which models you can use, where the data centers need to be hosted.
So I think that can also sometimes be a bit tricky almost when we are used to a certain AI model or workflow for our personal use or maybe from a previous client, and then we go to a different client and they say like, "No, you cannot use this one here.
It's not approved by our security department," or something.
So I think that's something to always keep in mind a little bit, that different clients have different, different rules and standards when it comes to how you can use AI. Yeah. W- what's your favorite tool right now? Mm.
I'm, I guess, an Anthropic fangirl in general, so I've, I've always liked Claude. I think, uh, Claude Code is great.
Uh, it's funny that we talk about it now because just before we hopped on this call, I, I think the internet is about to break because Anthropic, either they made a mistake on their website or they are going to exclude Claude Code from the pro plan, so from the $20 plan.
Um, so- No way... you would need to buy the, the max one, like the $100 one. Um, and everyone on the internet is like, "Ah, you cannot do this." So I find that very interesting.
I will, uh, I will keep an eye on that to see if that's actually happening. But while we have it, I think Claude Code is, is quite good. I think I've tried most of them.
I tried Cursor for a while, which I thought was good. Um, I have used GitHub Copilot too, which is not my favorite, to be honest, but I think,
yeah, I can, I can adapt usually to whatever, um, the team uses in a client project. For, for personal use, I prefer Claude Code. Yeah.
I think it would be a major difference, I don't know, going from, uh, Claude Code to Copilot. Like, completely two different worlds, right? Yeah. Yeah, yeah, yeah. I'd be very sad. I'd be very sad. Yeah. Yeah.
[chuckles] Cool. So, um, uh, you know, I, I'm a freelancer myself.
I do that as, as well, uh, among all other things, and one thing that I've been struggling with a bit more bef- like, uh, compared to before the AI, is how do I price my stuff, right?
But most of all, if you, if you work by the hour, now the typing is done by a machine, and typing was actually something that was time-consuming, were kind of include- Yeah... was included in, in, in the hourly rate.
So if, if now that we have the agents doing maybe the 80% of the job and we do the 20%, which was actually...
which is actually the most important part, so the signing the, the solution and making sure it's reliable, maintainable, and all those things there, what, what are client paying for now, in your opinion?
Yeah, that's a good question. I haven't really adapted my pricing strategy, I think, because I think clients are still a little bit slow maybe to adapt to, to this change.
I think what they are paying for really even more than before is just our expertise, our ability to really determine what it is that we want to build, and maybe our focus is shifting a little bit away from the focus being on writing the actual code and now it's more, like, prompting the AI to write the correct code.
But I think even before AI, the most important thing was to be very clear on what we want to build, to really make sure that we build something that's valuable to our users, and to be very clear on the requirements and all the boundary conditions, and I think that remains true whether we use AI for it or not.
Um, just how we achieve that maybe is changing a little bit. When it comes to pricing, I think one thing that many freelancers try to do is to offer, like, package prices instead of hourly rates.
I think that might become even more interesting now with AI, because you could say something like, "Hey, I'm gonna build you an app for, I don't know, 5K." And maybe a few years ago, that would take you a month to do.
Now you can do it in three days, so you can make your 5K in three days instead of a month, so that's obviously great. I think that's sometimes a bit difficult when the requirements aren't completely clear upfront.
So for something like that to work, you really need to have, like, a very detailed description of the features that you have to deliver and what's included in the scope.
So from my experience, that's often the most difficult part to really align on that with clients. The other thing is if you, if you stick to more like an hourly or daily rate,
I just think that how your day looks maybe changes a little bit. So maybe, maybe you start a couple of AI agents in the morning, and then you go for a long walk. Um, but that's...
Like, in a way, I think that's, that's part of our job now, you know? I think it's, it's completely fine to say like, "Hey, we are not writing every single line of code anymore."
Um, but to make sure that your agents are doing what you want them to do, that's part of your job, and I think it's completely fair to, to charge money for that. Yeah. Yeah.
Yeah, another, um, solution, another offering I've seen working for some people is kind of produ- productize your offer. Yes. Yeah. So they offer like a, I think a either a weekly or monthly fee, fixed. Yes. Yeah.
And then they can use you as you... can use a, can use you as a, as a Netflix subscription. So as much as you want, and you can cancel as- Yeah... uh, whene- whenever you want. Mm-hmm.
So I think in that case, yeah, it can be quite, uh, quite interesting. Yeah. Yeah, definitely. Um, yeah. Uh, was there ever a moment that, um, you know, your usual way of pricing kind of worked against you
because of how AI is working now? Mm. I don't know if that happened so much with AI. Yeah, I think in general...
So I right now still charge an hourly or daily rate, so kind of n- time and input based rather than output based, and there's, like, obvious problems with that pricing model. So
sometimes I have days where I feel like- Really bad about charging my client for a full day just because I noticed that, I don't know, maybe I didn't sleep well, uh, and I wasn't as productive.
But then also sometimes I have days where I'm like, "I was on fire today. Like, I achieved so much, I should actually charge them three times as much," you know? But I don't.
So I think in a way it balances itself out, uh, a little bit, like those slower days and the ones where you are a bit more productive.
But I think in general, yeah, the problem with, with time-based pricing is that, yeah, you, you kind of measure the input. Like, you measure how much time you put in. You don't measure the quality of the output.
The benefit for us freelancers is that we basically take zero risk. Like, no one, no one can come after us and say like, "Hey, you didn't deliver what you promised," because we delivered an hour of our time.
You know what I mean? So it's like very, very little risk. It's, um, it's pretty straightforward. We don't have to worry as much, I think, about liability and things. Um, but yeah, I definitely had a few
projects, I think, where I felt that my hourly rate didn't really reflect the value I was delivering. Um,
but yeah, I haven't really found a great way, I think, uh, to, to productize my, my services because what happens quite often is that clients will just hire me and put me in an existing team temporarily if someone is maybe on maternal leave or something like that.
And it's not so much about, "Hey, we want you to build this very specific product or feature." It's more like, "Hey, we want you to support our team." And then we kind of agree from week to week what I'm gonna work on.
So it's often really difficult to scope it out clearly what, what, like, a product package could be, and then it's even more difficult to put a price tag on that. Yeah, that makes sense.
Um, I think you touched upon a very important, uh, thing, like, that we as freelancers, most of the times we're hired to produce an output rather than an outcome. Yeah.
Um, and I think, I think we can touch upon this a bit later in the, on podcast. Yeah. Um, so sure. So, um, w- you started kind of talking about, uh, some of the challenges that freelancers have.
What, what's, in your opin- in your opinion, the, um, hardest thing about freelancing as a software engineer?
I think for me, what's always been a bit difficult is wearing all these different hats, and especially now with this whole, like, uncertainty around AI and politics and, um, economic crises and whatnot, um,
the ability to just adapt to change rather quickly. Like, these are two difficult things, I think.
I often get people who kind of reach out and tell me that they want to start freelancing in software engineering because at their current job, they have too many meetings or they spend too much of their time on non-coding tasks.
And I think with that, I'm always a bit careful because I think as a freelancer, you might have even, even more tasks that are not coding, you know, because you have to take care of your marketing and client acquisition and bookkeeping.
Of course, you can outsource all of these things, but the, the question is if you have, like, the, the resources to do that from the start.
So I think that's one, one thing to keep in mind, that as a freelancer, you are running a tiny little one-person business, and you are the head of marketing and the CFO and the intern all at the same time.
Um, that's an important thing to keep in mind.
And yeah, times are, are changing quite rapidly right now, and I think the freelancers that I see who are struggling a lot with that are really the ones who are super fixed on a specific technology or tech stack, or who are just used to work in, working in a certain way and who now have to adapt and they're not really, not really equipped to do that, I think.
So, um, if, if, um, we have to think about
a way to make our lives better, uh, with, for, for our business as a freelancer, because one of the things that I'm also struggling with, for example, is scalability.
So it's, it's, we, we're just selling our time most of the time, so we can't really, uh, we can't really clone us, right?
Even though, no, with, with, with AI it's a bit funny because I have, I could have maybe two more clients rather, uh, compared to before, and I am finding myself in this constant, constant like, um, switching between different things where I have maybe an agent working on one client, an agent working on the client.
So it, it's getting a bit, a bit stressful, right? So- Yeah...
um, I, I haven't still under- uh, cracked the code there, how to manage all these, uh, things happening at the same time within different scopes, in different, um, different projects.
Uh, but still, yeah, that, that surely is a, is a plus, right? Um, but, um, another thing is usually freelancers, they, as you were saying, they, they produce outputs. They, they're hired for that. Um,
what if we, we, we try to change that? What if, what if we start to focus on outcomes instead?
If we stop thinking about the, only the, the technical side and maybe start being a bit more active towards what the client is asking us. So they come w-with a q- with a, with a specific task.
We can, we can zoom out and say, "Look, maybe there is a better way to do this," and you know. So thinking about the impact we have. So do, do, do you see this working in your experience? I think it could,
but again, right now I feel like clients are still quite slow to adapt to this mindset shift. Maybe that's also because I'm in Germany and we are extra slow.
Um, but it feels like at least here clients are still very much in this, in this mode of like, "Show us your CV, tell us your hourly rate, and then we will hire the cheapest person with the most experience."
So I think it's gonna take a while to have this shift to have, like, a more outcome-based approach. I think it's possible. I think now is the, the right time to do it.
I also find your point quite interesting that now basically we could have two or three clients at the same time or more clients than before because all we have to do is, like, have different agents running in the background.
And on the one hand, I completely agree, but also there, at least for me, often I think my clients also pay me for my availability.
So not only for the code that I write, but they expect that on the days where I work for them, I reply to Slack messages within, like, a timely fashion, you know?
Um, and that becomes more difficult, I think, if you're, if you're trying to balance several clients or more clients than before.
But I think also in my head I'm a little bit like, my goal is not really to build more applications or produce more, more output. I think I'm more wondering right now how can I deliver the same value but in less time.
I just, I just want a bit more time for myself, you know? I think, I think the, the 40-hour workweek is, is just completely done. I think it doesn't, it doesn't match our, our current lifestyles and society anymore.
Um, I think something like a four-day workweek is long, long, long overdue, and I think that's what I'm trying to figure out. How can we, how can we maybe use AI to make something like that happen?
Like, do we really need to create more and more and more code, or is it just an okay co- okay goal to create the same amount of code but in less time and then we all have a three-day weekend? Yeah.
Yeah, that's a good point. Um, and you know, this makes me think about... So the, the, the trend I'm seeing in, in, in the industry right now, even for, like, em- employed people, like in, uh, in general,
is that we are using AI to produce more and more code. So you know, there are companies that are even, that even have leaderboards about who uses more tokens and stuff. It's, it's going very crazy, you know?
And, uh, I'm seeing that, that the, the main narrative is that people keep focusing on how much things we can produce rather- Yeah... than what's really useful about what we are producing.
So I was maybe a bit naive to think that this would have been, would have changed a bit so...
Okay, now we can write code faster, now we can take more time to focus on other more important things rather than just spit code, like, uh, constantly.
So I think for whoever freelancing, this would be indirectly also a side... like something that they, they will face because companies still think like that, right? Um, uh, so I don't know.
It's a, it's, it's a challenging task for sure to try to reduce that when the clients r- really want more from you. They still want a lot because they need to push, push, push, push.
And now with AI they think it's even faster and easier, right? Yeah, that's true.
I think one thing that's definitely become easier with AI is also just rapid prototyping because I think that's one thing that I've seen quite a lot, that we are just building the wrong thing and we don't realize until it's too late.
So I would, I think, absolutely advocate for just vibe coding and a first MVP in a day or something and getting it out there and getting user feedback and then deciding if it's worth it to focus on that more and to build it out into an actual production-ready application.
I think that has become a lot easier. But then again, if you think about, like, bigger corporation, the speed at which we write the code, that's not always the only limitation, you know?
There's, like, all these, like, big processes in the background for when we can ship things and when we can set something live.
So sometimes it's like the code is ready, the application is ready, but the company is simply not. So that's the other thing.
We also need to change these processes to, to be a lot faster, I think, um, if we really want to speed up the, the process of product development a little bit. Yeah. Also to me it feels like the...
Like e- even before AI, I think the, probably the freelancers that probably were doing best are
the ones that, um, bring a bit more to the company, like maybe some product thinking, something, a bit more communication skills and, um... So not, not just, you know,
executing specs but maybe like even pushing on specs. I think that's gonna become even more important, especially things like communication skills.
I think this, mm, this kind of mindset shift where everyone is like, "Ooh, AI is going to take our jobs," to a certain degree, yes, but I think the question is, like, which, which part of our jobs?
Like writing the code- Sure. I think that's, that's going to happen. Um, but what are the parts of our job that AI can't do?
And I think that's really the human side of things, you know, like being empathetic and focusing on clear communication and just, just being like a nice and kind person to work with.
I think that's gonna become even more important. I think for me, that always has been quite important.
That's, that's been so far how I found all of my clients, so only through like word of mouth referrals and my network, because I'm honestly not the best developer. Like my CV is completely average.
Like I can write code, sure, but there's, there's always people out there who write better code than me and charge less for it. That's, that's always going to be the case.
But I feel like when people work with me, they want to work with me as a person. And also when, when I get feedback on a project, it's never like, "Ooh, like your, your React code is so clean," or something.
It's really more about being proactive and being a good communicator.
So I think these things will just become more and more important because I think our, our technical skills will no longer be able to, to set us apart from, from the masses and especially not from AI. Yeah.
Well, th- this could be a, a good idea, right? So y- you're selling your, your brand rather than- Yeah... like yourself, rather than the, your, your skills. I mean, they're still- Yeah... important, but yeah, right.
It's a little scary, honestly. Like it, it requires like I think a, a fair amount of being willing to be very open and vulnerable on the internet or wherever it is you, you do your, your client acquisition.
For me, it has worked quite well, but it's, it's a little scary sometimes, um, because I think also it's a little polarizing.
Like once you start to really put yourself out there and try to build a personal brand that is really personal and authentic, then you're gonna attract some people, but there's also gonna be other people who will say like, "No, I absolutely do not want to work with you because that's not what we're looking for."
So that can also happen, but I don't even think that that's such a bad thing.
Like maybe, maybe it's actually a good thing to, to only attract the r- the right clients in a way who value you for the, the personality that you bring to the team. Yeah.
Yeah, because they're probably the ones that have been following you, they read your content, so they, they like who you are and they want to work with you for that reason. Exactly. Right.
And I mean, they're, they're gonna get like a taste of my personality anyway once we work together. So I'm a bit like, why, why hide it?
Like if they, if they can't really handle the way I communicate or my slightly dry sense of humor, then we might have a problem working together anyway, so you know, why not just get that out of the way?
And, um, I have some people who reach out to me because of that, and then I'm sure there's a bunch of people who don't even consider working with me because they've seen me on LinkedIn or something and they're like, "No, she...
[laughs] Like we, we don't wanna work with her." And that's completely fine, I think. Yeah. When, when did you... When, when did your, let's say, personal brand started to work with you? W- w- what...
When did you see something moving? Mm. So I started that maybe like, hmm, three years ago, I wanna say.
So I went self-employed three years ago, and I think, think that's when I started being a bit more active on LinkedIn. So what I did is just
w- something like a building in public approach where every week I kind of documented how my journey into freelancing was going, like how I was fighting German bureaucracy and, um, what like my current obstacles were.
And that seemed to, to resonate quite well, and I really noticed that I was building a community online, and that's something I was really craving because I didn't really have many self-employed or freelance people in, uh, in my circle of friends or, or family.
So I needed that community a little bit. Um, so yeah, that's when, when I kind of started that. And to be completely fair, I think that was exactly the right time to be very active on LinkedIn.
I think especially LinkedIn has changed a lot since then. I'm much less active these days. Like every once in a while I try to make a bit of a comeback, but I'm also slightly annoyed with it.
So it's often like I open LinkedIn and I close it again because I can't take all the AI slop.
Um, that just, yeah, as a kind of a disclaimer, I think these days it's much harder to, to grow on LinkedIn because the algorithm has changed and the platform has changed.
But three years ago it was not that many people were actively posting, so if you did, it was, it was kind of easy to, to grow a following there quite, quite quickly.
Um, so I did that for a while, and then at some point I had this, this huge network, um, which, yeah, sometimes turns out to be quite, quite helpful. Yeah.
Yeah, it's definitely a full-time job to stay up to date with the, all the changes in the algorithm and everything. Yeah. Yeah. That's, that's it. Yeah. And I think for a while
I tried to optimize a little bit for the algorithm because I was like, "Okay, if I'm, if I'm gonna invest time into posting, then I wanna make sure that it reaches the right people."
But at some point I just couldn't be bothered anymore to follow all these unwritten rules for an algorithm that is, is essentially a black box. Like we don't know how it works. It's...
It can be changed at any point, and we won't notice. Um, so now I'm a bit like I, I post whenever I want to. I don't really think about it too much whether a post is going to perform well or not. Um- Yeah, but it's...
That's the thing, like if we, if we really want to take it very seriously, then unfortunately this algorithm is something we, we have to consider, and I think at some point I just decided for myself that I, I didn't wanna do that anymore.
Um, yeah. Yeah. I, I, I have my own AI agent that does all this for me. [chuckles] Ah, amazing. So he, he knows ex- he knows exactly- How, how do you use it? Like, what does that look like, your AI content flow?
Well, um, so th- this, this, um, agent knows exactly, based on some knowledge I found online, uh, how the algorithm works right now, and I up- I update the knowledge base constantly. So I have...
Every time I want to re- create something on LinkedIn, he knows, uh, what, what to look for, what, what can... what works, what doesn't work, what strategy I should use. So it's more like a,
yeah, like a consultant, LinkedIn consultant- Mm. Yeah, yeah... I can use anytime. Please. And that's been very useful. So because I don't have to be reading every, every time what, what I have to do, the guidelines.
I just ask, "Look, I have to write this post about this and that. What do you think is gonna work better?" And- Mm-hmm... to be honest, so far it's been quite good. It's been working very well. Yeah. Nice. Yeah.
That sounds good. Yeah. And, and this brings me to another, uh, thing, which is, you know, how can we use AI to support our business, right? So for example, I'm doing this for, for LinkedIn- Mm...
but it could be also that we can use it for outreach or for creating more content or for project management. I don't know. Um, so the, the, the way, the way I like to, to see agents is like my employees. So I'm, I'm...
I've started like having a look in OpenClue, but it didn't really convince me, so I started to build my own, my own crew of agents. Mm-hmm. And one of them even is, has its own email address, so that, that's good.
[chuckles] Mm. But, um, it sounds, sounds, sounds quite useful. Like do, do... Are you, are you using AI in this way so to support your business somehow, and is it been useful?
A little, but I think not quite as, as professionally [chuckles] as you just described. So I think right now my favorite way to use it is really just for brainstorming, because that's the, the other thing.
Again, we're like, as freelancers, we're, we're a one-person business, so we don't really have other people to bounce ideas off. So I think that's what I've been using it for a lot, to just...
Like these days I'm thinking a lot about how I can move away from just selling my time, you know, how I can, uh, either find like a different way of packaging my work to move to something more like a productized service, or even find different ways of, of making money.
Like what, what happens if AI really takes our jobs, you know? Like what, what are we gonna do? So just to kind of brainstorm these ideas and, um, have
s- someone, an AI robot in that case, to, to challenge your ideas a little bit and map out a plan for that, for what that could look like, I think that's very helpful.
Um, these days I'm starting a mailing list, so I've been using it for that quite a lot, um, just to, to set up like a website for that. It was, it was super quick. Like it was just done in an hour and a half.
And also to kind of find my, my angle and my voice for what I want to write about.
Like what are my, my influences, what are the, the sources I consume, um, and how can that influence the, the content that I want to write. For that, I, I found it extremely helpful.
When sometimes you just feel like you're, you're running out of ideas, like you, you want to share something, you want to write something, but you have a little bit of writer's block.
Um, I'm not necessarily saying like use AI to write your content, but use it to, to give you ideas, um, that you can then write about. Yeah. Yeah. Yeah, the, the famous writer block. Yeah.
Um, do, do, do you know Zettelkasten? Zettelkasten? Yeah. It's a like note-taking technique. I heard about it. I never used it before. Okay. Yeah.
That, uh, w- when you said writer block, it made, made me think about that because it's, uh, it's quite good for it because the way you write notes,
eventually they kind of, uh, you can just take them, merge them together, and you can always use them to create anything, an article, a book, or anything you're writing. Yeah.
Uh, because the way you write them, it's, it's like a, a standalone paragraphs or chapters in a, in a, in a book, so you can just mix them up. Ah, yeah, yeah, yeah. It's pret- it's pretty interesting.
It's really interesting, yeah. Yeah, that's nice.
I've also noticed that it really becomes easier the more I write, the more I post, the more ideas I get, which is also why right now it's difficult because I haven't been active on LinkedIn or anywhere else at all.
So then sometimes I'm sitting down and I'm like, "Okay, today is the day I'm making my comeback to the internet. What will I write about?" And I have no ideas.
But when I was super active on LinkedIn, I would get like so many more ideas just from the conversations I had on there.
Like, I would post something and just from the comment section alone on one post, I would get 10 new ideas for other things I could write about. So it's, yeah, it's a bit of like a snowball effect, I think.
Um, and I think it's also just a, just a habit and a routine to, to get into writing again a little bit more, and then I, I hope that it will feel more natural again at some point. Yeah.
I mean, while, while your agents are working for you, you can write, and that's also possible now, right? Yeah, exactly. Yeah. Cool.
Um, so let, let's say- Five years in the future, um, how do the top 10 freelancers look like? Do they focus on... How, how do they price themself? Hours, outcomes? What, what, what do you have in mind? Ah, good question.
I'm even... The thing is in, in my head, like, the term freelancer is really, is really kind of someone who sells their time. So
what I'm thinking is, and I have kind of like a love-hate relationship with that term, but I think we will all really move more into, like, a consultant role, where
our job is to stay up to date with the latest developments in, in tech in general, in AI in particular, and just advise clients on how, how best to use it, um, and just support them to adapt to these changes quickly so that they can continue building their, their products.
Um, that's kind of how, how I see it. Honestly, I think 5 to 10 years with the, with, like, the rapid developments we're seeing right now is a crazy long time.
Like, I can't even tell you what my job is gonna look like in a year or two, so I think it's gonna change, and it's gonna change quite, quite fast. Um, I also...
I mean, yeah, you, you call this the, the Product Engineers podcast, right? So, um, I think we will also, that's my guess, all more move a little bit into, like, something like a product engineer role.
So these days we still have, like, we have front-end developers, and we have back-end developers, and then we have DevOps or something.
And even within the front end, there's people who call themselves React developers or Vue.js or something. I, I think that's gonna get a bit more broader in a way.
So since with AI, we, we, we can all build anything basically, um, I think it will become less common to have these, like, super niche-y specializations where we're saying like, "Oh, I am an expert for C++," or something.
I think it will really be more important to just be able to build, to build cool products, um, and be able to, to choose the right technology, but then you don't need to be an expert in the specific m- programming language to, to do that, to implement that.
Yeah. You know, it's been always quite challenging for software engineers in general, freelancing or not, to, um, relate their work to an outcome. I mean, even the, you...
Maybe you're working on a ticket, and then you ship the thing, and then you don't know what happened, right? Yeah. You, you don't know the impact you had. A lot, um...
Sometimes it happens, but I think generally that's not the case. Yeah. Uh, someone has to check. You know, they don't check at all.
So I've been trying to look at other, uh, professions, other roles, like maybe product managers, how they, how they do it when they go consulting, so.
And they, they can work a bit more on, on, on the, on the value they bring and on the outcome because maybe it's about bringing this metric up or this, you know, im- improving some kind of, uh, uh, usability issue, you know, things that you can measure.
So yeah, I think, yeah, as freelancers, if, if we can get in that direction where we s- we can prove that we, we moved the needle somehow, that would be quite good.
Also a way to maybe to, to, to not bill by the hour anymore, let's say. Yeah. Yeah. Yeah, definitely. Yeah. So
let's say before wrapping up the episode, uh, if there is a senior engineer that is listening right now that wants to get in, into freelancing, uh, what, what, what would you tell them to do before quitting their job?
What would be the most important thing? Um, it's not... Like, it's a bit unsexy, but I think just get your finances in order. Like, run the numbers [chuckles] and see if you can actually do it.
Like, unless you already have, uh, a pipeline of prospective clients, um, I would say really make sure that you have enough runway, especially, like, in these times right now.
So before you just, just quit your job, make sure that you have enough savings to, to survive for a couple of months. Um, that of course really depends, like, on your, on your living situation.
And, um, just be realistic about how, how much you can charge. Um, so really do your, your market research. Speak to other freelancers in your area, um, in your field of expertise. See what they are charging.
Um, because I think many people, the way they go about pricing is they look at their costs and then try to figure out how much they need to charge to make ends meet. But that's just not how the market works.
Like, the market determines how much you can charge, so you really need to figure out your value, what you can, what you can deliver, how much you can charge for that, and then see if that's enough to, to cover all your costs.
Um, so that's a big one, I would say. And then really my advice, I think, would be to try and activate your network.
So if you're already working in, in tech, um, try to reach out to, to people you've worked with in the past. Let them know you're opening the doors of your freelance business.
Maybe they have little bits and pieces of work for you, and even if not, they might be able to refer you to someone else, and I think that's always solid gold. Um, what about platforms like Upwork or, or similar?
Um, th- they've never worked for me. Like, I, I've tried a couple, not, not Upwork, but, like, smaller, more local ones, um, here in, in Germany or Europe. I'm registered with a, a few of them, but I have...
I've never found, like, a single project through them. Like, I... They sound so nice, like the concept to me sounds super nice. I will just upload my CV, and then clients will come to me.
But at least for me, that's not how it works. But I think that's, again, the point of, yeah, my CV is just not that special.
So I think it can work really well if you have a lot of experience, mm, especially if you have a lot of experience with, like, a technology that's very, like, sought after.
So I think right now, um, one of the hottest things is still DevOps. So that's, I think, one of the things where it can work really well.
Like, at least here in Germany, um, you can, you can charge a lot of money for that, um, and probably won't have, have trouble finding clients.
But especially if you do something super mainstream like I do, you know, kind of like front-end or full stack web applications, everyone knows how to do that.
Um, and there's always gonna be someone who can do it a little bit better and a little bit cheaper. So that's why that hasn't really worked for me. I think I really need to talk to people.
I need to get them on a first call. I need to get them on the phone, and then they can decide if they wanna work with me or not.
But if all they see is, like, my profile on a platform like, like, uh, like Upwork, um, that's... that hasn't really worked for me. So also maybe build your personal brand. That would be also very- Yeah...
very good movement. Yeah. Takes time, but- Yeah... yeah. It takes time. Yeah. That's the thing. Yeah.
Like, I think for me that only worked because, as I said, I was really, I was really craving to feel as part of a community. Um, that's kind of why I did it.
And then the f- fact that I got some projects out of it, that was more like a nice side effect.
But I think some people start posting on the internet or start building a personal brand with the goal of finding clients, and then they're disappointed if nothing happens after a month.
But I think if you, if you can kind of try to find a different angle and try to make it enjoyable and try to get something else out of it too, then it really makes it a bit easier to, to stick with it in the long run.
So, Judith, if, uh, anyone wants to ask you any questions, where, where can they find you outside LinkedIn? Yeah, LinkedIn is a big one for sure.
Uh, I think the best place to go right now is my website, judithbuhler.com. We can put that somewhere in the notes maybe. Yes.
Um, these days I'm building out a mailing list, so I'm currently starting to write the Low Noise newsletter, so that would be the best place to go. And the goal is just to,
to write about finding a more calm approach to working in tech. I think that's really my main thing right now. As I said, I don't wanna... I don't necessarily wanna produce more and faster.
I just want to find a way that we can all stay in tech and enjoy working in tech, um, without, like, burning out and without, um, hustle culture and one productivity hack after the next. Yeah.
I think, I think I need to read it, so looking forward to it. [laughs] Yeah. Nice. Okay. Um, thank you very much, Judith, to be on the podcast. It was nice talking to you and- Yeah... yeah. Thanks for having me.
It was great. We don't wait for ticket. We don't wait for specs. We build it, we own it, we ship what comes next. Welcome to the Product Engineers podcast.
Wow.