Alex Klimenko (@aek15) header
Alex Klimenko (@aek15) profile picture

Building AI for industrial equipment @Soply. Edge AI, robotics, CAD. Lifting, endurance sports, travel, game dev.

Amsterdam, The Netherlands
Joined April 2009
Photo by Alex Klimenko (@aek15)Photo by Alex Klimenko (@aek15)Photo by Alex Klimenko (@aek15)Photo by Alex Klimenko (@aek15)Photo by Alex Klimenko (@aek15)Photo by Alex Klimenko (@aek15)Photo by Alex Klimenko (@aek15)Photo by Alex Klimenko (@aek15)Photo by Alex Klimenko (@aek15)Photo by Alex Klimenko (@aek15)
Lots of new followers lately, so here's the short version of my story. 19 years in tech. Dozens of stacks and technologies. Yet every day, month and year looked exactly like the one before. So I left a very awesomly-paid job and went traveling. I wanted freedom but had no idea what to do with it. Turns out the answer was work where I use everything I know and learn something new daily. I haven't had two similar days since. We build tools for industrial machine builders and manufacturers. It comes with its own struggles. 19 years in tech taught me how to build. It didn't teach me how to sell. Expect lessons learned the hard way, robots and industrial automation stories, complaints about tech, and occasional attempts at humor.
8
1,286
Back home
2
7
158
Hot take: B2B founders should never post their founder journey on LinkedIn. Your buyer isn't rooting for you. They're trying not to get fired for picking you.
3
4
235
I spent a week in Georgia 🇬🇪 working with this view. But I miss Amsterdam already. Going back home tomorrow.
1
2
150
Friday 13th. Evening. We just finished rebuilding some core functionality in the product that involves multi stage data migration. It worked perfectly on test env. Nothing can go wrong, theoretically. Would you proceed today?
1
128
Anthropic Launches Claude for Manufacturing. This update made me feel quite uncomfortable tbh and it took some time and very long shower to realise that it does not change a lot for us. It will just kill the part of the product that we decided to kill already. But the fundamentals stay.
2
4
401
Built on my internal admin page a cumulative status reports from all external tools the product uses. Just one prompt asking Claude to list all services we connected to and put their status pages together.
2
174
Raise or bootstrap? Do not listen to anyone. It will be a right decision only if it is yours. Founders argue about this all the time. The case for bootstrapping is pretty convincing: building is cheaper now, the tools are there, and you don’t need much money to get a product out. Yes. But your competitors have access to the same AI and the same tools. You still have to get buyers to notice you. People tend to buy from companies they already know, and getting in front of them takes time and money. That’s where funding can give you a real advantage. “Then fund it with customer revenue.” Sure, if the timing works. But what if your sales cycle is 9–12 months and you’ve got six months of runway? You still have bills to pay while you’re waiting for those deals to close. You can cover the gap with savings. Plenty of founders do. But draining your own bank account isn’t automatically a better business decision than raising money. There’s no single right answer. It depends on what you’re building, what it costs to reach your customers and how long it takes to get paid.
1
86
I took this screenshot pressing volume up - power. You can’t teach an old dog new tricks.
116
II pay MongoDB Atlas $150/month. Self-hosting the same setup would cost me $90. So I'd save $60 a month. Sounds like an easy win until you write down what that $60 actually buys: The math, per environment: 3 VMs: $30 VPN tunnels: $10 Backups: $5 Times two for prod and dev = $90 What becomes my job: - automatic failover - point-in-time restore - version upgrades and OS patching - monitoring and alerting - encryption at rest - restore testing That's $2 a day to not be my own DBA. Even if I spend just half an hour every month on these tasks comparing to my founder's hourly rate my TCO will explode. Business should be lean.
3
237
It feels at this moment that instead of telling AI to do things. I do what it tells me.
2
109
Today the Grok bot sent me a video of a competitor demoing their features in the same space where we are developing our products right now. I have not watched it myself yet, but I gave it to Claude to identify what we are missing and then write and prioritise these features on the board. Technically nothing is stopping me from typing another sentence and having everything built by next morning. Except there is no point in doing it earlier than when somebody asks for one of those. But if my approach was different I could have excluded myself from the loop and had one agent monitoring competitors updates and another building the features we missed.
3
130
When I am traveling I much less productive, but I have so many new ideas. If there was a way to leverage this.
6
6
480
I built a secret game room on my personal website. And visitors can play Sonic there. Looks very cool. I recently started something I’ve been putting off for a long time: making games and I will use this space to keep it separate from my very serious main page.
1
2
139
I needed some specific functionality from the text editor I use (Sublime) and instead of searching for it and installing myself I asked codex to develop one from scratch and it did. I haven't touched a single file with code and it works as a charm.
3
4
283
AI agents already out-code the average engineer. So should we rethink testing from scratch?
2
1
169
Magic Tbilisi invited me for the third time in the past year.
1
114
Let's re-connect after the holidays period is coming. And this year I am planning to take the holidays break myself.
1
2
193
A startup becomes a real business when you have repeatable processes. But now everything is changing so fast. I am not sure if it is an accurate definition anymore.
98
Mechanical engineering + CS + MBA. Worked at top tech companies. Still can't calculate my odds of raising from VCs. Jokes aside: every founder should know the basics of finance, strategy, decision-making under uncertainty, people, communication and supply chain. Doesn't matter if you get it from an MBA or from Claude.
I redid the startup founder signals that lead to top tier VCs investing in the seed / A, after incorporating some feedback. This added ex-frontier lab and cyber security near the top, IIT as a positive, and educated in China as a negative. The main changes were; - extended founding year to 2019-2026, ~2x'd the number of startups - unified and added some signals Here are the new signals:
123
All weekend watching documentaries about deindustrialisation of the US. A good and a bad thing about the US is everything is happening so fast there. And when things are bad it is very painful for normal people.
119