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.
36 articles from this blog
Armin Ronacher experiments with Pangram, an AI text detector, by using an LLM to recreate a tweet and analyzing the results.
Armin Ronacher's critical take on AI 'doom' narratives and the push to pace frontier AI development.
Armin Ronacher discusses why anger at work is counterproductive, especially in tech, and suggests curiosity and excitement as better responses to AI-driven uncertainty.
Armin Ronacher discusses how LLMs are making programming language choices less consequential, leading to more developers using 'hard languages' like Rust and Zig for fast, small software.
An analysis of AI reasoning traces, how they are hidden, and the role of system prompts in reasoning effort.
Explores how AI-assisted programming affects coordination in large software projects, using the Tower of Babel as a metaphor.
Analysis of newer AI models producing worse tool call accuracy, specifically Claude models inventing extra fields in Pi's edit tool.
Exploration of coding agent loops beyond simple prompts, discussing harness-level loops and challenges with AI-generated code quality.
Analysis of US export controls on AI models, arguing they create a dangerous division based on nationality rather than safety.
Analysis of how large tech companies manipulate narratives around openness and access in AI, discussing EU's DMA and Apple's restrictions.
Armin Ronacher discusses how communities formed around opposition to something can lead to mob behavior, using LLM-skeptical developer spaces as an example.
Armin Ronacher explains why he uses 'clanker' instead of 'agent' for AI tools, emphasizing machines as tools without agency.
Armin Ronacher reflects on using AI agents (Pi) to build Pi, highlighting challenges with AI-generated bug reports.
Explores the challenges of using local AI models for coding agents, focusing on usability gaps like tool parameter streaming.
Analysis of inflated word usage in coding sessions, linking LLM influence to Google Trends increases.
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.
