Kitaru
Kitaru is open source and ready to use
Kitaru is live: open-source infrastructure platform for running Python agents in production.
Read post → Blog 
Thoughts on durable execution, agent infrastructure, and building reliable AI systems.
Agents 4 min read
The Anatomy of a Production Coding Agent
A production coding agent isn't a prompt and a while loop. It's eight stages, each with different failure modes, costs, and human touchpoints. Here's the full pattern.
Philosophy 4 min read
From Pipelines to Agents: How Orchestration is Being Rewritten
ML pipelines were DAGs. Agents are loops. The orchestration layer that worked for training jobs doesn't work for autonomous systems, and the industry is scrambling to catch up.
Philosophy 5 min read
From ZenML to Kitaru: Why We Built a New Product
We spent five years building ML pipeline infrastructure. Then agents showed up and we realized the next problem needed a new tool — not an extension of the old one.
Infrastructure 2 min read
Your Agents Need More Than Just Traces
Tracing shows you what went wrong. But what if you could go back, fix the input, and resume from where it failed — without re-running everything?
Design 3 min read
Why Kitaru Doesn't Use Journal Replay?
Every durable execution engine today forces your code to be deterministic. Kitaru takes a different approach — and it matters more than you think.
Infrastructure 2 min read
Why Your AI Agents Need Durable Execution
AI agents fail - they timeout, hit rate limits, crash on bad API responses. Without durable execution, every failure means starting over from scratch.
Agents 3 min read
Your Agents Are Not Microservices
Durable execution engines were built for payment flows and order processing. AI agents need something different. Here's why.