---
title: "Founder's Notebook — In the AI Age"
description: "Decisions, experiments and tradeoffs from building a company as a technical solo founder."
url: "https://owlmeans.com/blog/category/founders-notebook/"
lang: "en"
---

# Founder's Notebook

[← Back to blog](https://owlmeans.com/blog/)

-   [![How to fix already outdated product just after release in a week](https://owlmeans.com/_astro/hero.DIj7bXvB_Z11jeIa.webp)](https://owlmeans.com/blog/uroboros-engine/)

    [agents](https://owlmeans.com/blog/tags/agents/) [pipelines](https://owlmeans.com/blog/tags/pipelines/) [context-engineering](https://owlmeans.com/blog/tags/context-engineering/)

    [## How to fix already outdated product just after release in a week](https://owlmeans.com/blog/uroboros-engine/)

    Why the story-by-story engine behind OwlMeans needs a different unit of work, how pipelines and agents already cooperate, and what still separates us from developing the Platform with itself.

    Nov 13, 2026 Updated on Nov 15, 2026

    [Read more →](https://owlmeans.com/blog/uroboros-engine/)

-   [![I spent a month trying to replace our coding models](https://owlmeans.com/_astro/og-neutral.DLWcNA8V_P5qBf.webp)](https://owlmeans.com/blog/open-weight-5x/)

    [open-weight-llms](https://owlmeans.com/blog/tags/open-weight-llms/) [anthropic](https://owlmeans.com/blog/tags/anthropic/) [openai](https://owlmeans.com/blog/tags/openai/)

    [## I spent a month trying to replace our coding models](https://owlmeans.com/blog/open-weight-5x/)

    I tried open-weight models in our development pipeline for a month. Reasoning limits, response formats and provider behavior made the switch more work than I expected. These are notes from June, without a cost benchmark.

    Jun 22, 2026 Updated on Oct 15, 2026

    [Read more →](https://owlmeans.com/blog/open-weight-5x/)
