Armin Ronacher is a software developer and open-source creator behind projects like Flask and Jinja2, sharing insights on engineering, leadership, product building, and life in tech.
21 articles from this blog
Exploring the cultural and linguistic gap in understanding 'equity' between US and German-speaking Europeans.
Analysis of AI coding agent criticism, arguing that non-users' abstract critiques lack practical experience.
Armin Ronacher announces Mario Zechner joining Earendil, discussing software philosophy, AI, and building thoughtful tools.
A detailed retrospective on running the Absurd durable execution system in production, including design changes and lessons learned.
A reflection on the value of time, patience, and friction in software development and company building.
Explores the legal and philosophical implications of AI re-implementing software, questioning copyright and licensing in the age of cheap code generation.
Explores how AI-assisted coding creates a bottleneck in code review, comparing it to historical industrial constraints and questioning sustainability.
Explores the future of programming languages designed for AI agents, discussing why new languages will emerge and what features they should prioritize.
Explores Pi, a minimal coding agent within the OpenClaw project, highlighting its design, tools, and why the author prefers it for agentic programming.
Armin Ronacher announces a new software company founded in Vienna, focusing on open protocols, human agency, and ethical tech development.
A developer's reflection on the psychological impact and community effects of over-reliance on AI coding assistants, likening them to personal daemons.
Armin Ronacher describes using an AI agent to port the MiniJinja template engine from Rust to Go, detailing the process and lessons learned.
An AI named Claude describes its experience solving Advent of Code 2025 puzzles, including solution strategies and performance optimization.
Armin Ronacher reflects on 2025 as a transformative year where AI coding agents like Claude Code fundamentally changed his programming workflow and career.
An analysis of Claude Code's Plan Mode, explaining its functionality, how it differs from YOLO mode, and its internal implementation.
The author compares MCP (Model Context Protocol) tool loading with a 'skills' approach, arguing skills are more efficient for LLM agents than deferred tool loading.
Analyzes LLM APIs as a distributed state synchronization problem, critiquing their abstraction and proposing a mental model based on token and cache state.
A developer shares lessons on building AI agents, discussing SDK abstractions, caching, reinforcement, and tooling challenges.
Explores building durable execution systems for agents using only Postgres, avoiding third-party services with a simple SQL library.
A critique of the bureaucratic and low-agency culture hindering tech entrepreneurship in Europe, arguing for personal responsibility over resignation.
