Stop Allocating Nothing: How We Tripled Java TreeSitter Performance...
(...on our CI/CD performance regression tests)
Upgrades in this post are
Quantifying the Tier List
We've released three Power Rankings now, and since the beginning
GPT 5.4, Ranked
OpenAI released GPT 5.4 yesterday, so we took the opportunity to
The 26.02 Coding Power Ranking
For 2026, we’ve split the Power Ranking up into Planning and
What's new in Brokk 0.23
The headline feature in Brokk 0.23 is that we are making
Your ChatGPT Sub in Brokk - LLM Reasoning with Context-Aware Dev
Large Language Models typically process code in isolation, lacking visibility into your
Code Review: the final boss of AI-assisted software development
As models become smarter and coding tools become better at harnessing that
New Year, Clean Branch
Starting a new software project is often compared to the initial git
2025 The Year AI Coding Grew Up
2025 marked a fundamental shift for Brokk.ai, moving to a true
Future Dev: The Three Pillars of Brokk
As software projects grow in scale, traditional developer tools often struggle to
Why Gemini 3 Flash is the model OpenAI is afraid of
When Gemini 3 Pro Preview launched, Sam Altman called a Code Red
The Reliability Gap: Managing Expectations in a Multi-Model World
For engineering teams prioritizing privacy or cost control, the allure of self-hosting