About Building at the Edge

Building at the Edge is a personal blog about thinking clearly while building software, teams, and AI systems under real-world constraints.

I run this blog because writing helps me think.

It forces me to slow down, examine decisions after they’ve been made, and make tradeoffs explicit. Over time, it became a way to capture lessons I would otherwise forget — things that only become obvious once a system ships, breaks, or meets reality.

Where I’m Coming From

I work at the intersection of research and real-world systems.

I’m a co-founder and CTO at Apoco, and I work as a tech lead with IBM Research on applied R&D. Over the years, I’ve built and shipped systems across robotics, IoT, and AI.

Since 2022, my main focus has been agentic AI: what actually works, what breaks, and why — especially once systems leave demos and meet production constraints.

I’ve studied and worked across different environments and organizations, which taught me a simple lesson that shows up repeatedly in this blog: constraints change, but the tradeoffs don’t.

What You’ll Find Here

This is not a news feed and not a collection of frameworks.

You’ll find:

  • Essays when something feels worth slowing down for

  • Shorter notes written in the middle of building or experimenting

  • Practical lessons drawn from real systems, not abstractions

The intent is clarity, not completeness.

Access and Expectations

Everything on Building at the Edge is free. There are no paid tiers, pledges, or gated content. If that ever changes, I’ll be explicit.

Posting cadence is intentionally light: roughly one longer post per month, plus occasional shorter notes in between when useful.

About Me

I’m Matous Havlena.

If you want a more formal background or current role details, you can find me on LinkedIn or at matous.havlena@gmail.com.

I’m always happy to hear from readers, especially when something here helped clarify a problem you’re dealing with.

Thanks for reading.

User's avatar

Subscribe to Building at the Edge

Clear thinking on building software, teams, and AI systems under real-world constraints.

People