Philip Heltweg

A look back at our initial 2026 plan, what changed while building isselthal industries, income and expenses for the start of 2026, and why we now focus on agency work.

In July, after more than six years in 'TODO'/'DOING', I could finally mark my longest running task as 'DONE'. As the title spoils, it was about my goal of getting a doctorate.

How to build a local evaluation pipeline to measure, compare, and continuously improve the quality of your AI workflows, without relying on gut feeling.

After crawling and cleaning technical documentation locally, we add page classification, embeddings, and a knowledge graph to make the data much more useful for AI coding agents.

Join me in an async, email-based book club reading Designing Data-Intensive Applications, 2nd Edition. Inspired by Phil Eaton, we'll regularly discuss asynchronously via email and not have no video calls. Every level of experience and participation is wel

Small gotchas and learnings from crawling technical documentation. Saving documentation locally is helpful in your software engineering projects - for example, for local AI to offer your coding agents to use the documentation of software you use, or just

We are self-hosting LaSuite, an open-source Google Docs alternative, for our small agency - and opening it up for others to try.

For our recent work, we looked into how to consistently crawl technical documentation and make it available for further processing by LLMs or agents. Doing so would either allow you to use the documentation for downstream processing, or you could vendor t

I always wish my feed was more interesting to read, with more honest, raw posts and less pure announcements and AI slob. So, trying to be the change I want to see in the world, I will try to post more openly about trying to build an independent business (