What we've been building with AI
Real projects, real tools, real results — see how Codeminer42 is applying AI across client work, internal tooling, and R&D.
What We've Built
Privacy Skills for Rails
Three agent skills for Rails privacy compliance: assess your full codebase against GDPR/LGPD, apply privacy-by-design patterns to new features, and review changes before committing.
Sherlock
Autonomous investigation loop for finding and fixing performance issues, flaky tests, memory leaks, and reliability problems. Just point it at a symptom and let it dig.
Kanario
AI-powered thumbnail generator that creates consistent blog cover images in ~60 seconds using Claude and Qwen. Built in 4 days with TypeScript, no build step.
AI Agent for Talent Matching
How we built an LLM-powered agent for resource allocation, matching candidates to projects based on skills, experience, and team needs.
Agentic Engineering
Our methodology for working with AI coding agents. Not just using copilots — a deliberate engineering practice for getting the most out of AI-assisted development.
Rethinking Engineer Evaluations
We changed how we evaluate engineers — candidates now use AI during assessments. We evaluate how you use it, what decisions you make, and whether you truly understand the code.
Want to know more?
We're always happy to chat about what we've been learning and building with AI.
Let's Talk