Blog
Writing about reliable AI systems.
Essays, implementation notes, and practical guidance on replay, mutation testing, CI gates, and the operational realities of shipping agentic software.
Featured article
EngineeringMarch 25, 20266 min read
Chaos Engineering for AI Agents: The Case for Planned Failure
AI agents fail in ways traditional QA never catches. Here's what chaos engineering means for tool-using systems — and why it matters before you push to production.
March 18, 2026
5 min read
Reliability
Why Load Testing Misses AI Agent Failures
Load tests check throughput. AI agents fail on correctness. Here's the gap — and a better testing model for tool-using systems.
Read article
March 10, 2026
7 min read
Tutorial
Instrument Your OpenAI Agent in 5 Minutes
A step-by-step walkthrough: add Sepurux to an OpenAI agent, record a trace, run a reliability campaign, and see results in the dashboard.
Read article