Infrastructure for self-improving AI agents
Auto-tune your agent harness and prompts to reach production accuracy in minutes, not months.
Why Tellurio
Designed to help AI teams
maximize agent output quality
Defined in code
Define your agent harness architecture and prompts with Afnio SDK. Keep everything in sync in your repo.
123456789101112131415161718191234567891011121314Built for quality
Optimization of agent's harness and context using state of the art algorithms.
Self-Learning Agents
Production data used for continual agent optimization and learning new tasks.
Built-In Observability
Integrated logging and metric tracking into every module and function.
Products
Built from the elements up
to power agent development
Tellurio Studio
The hosted platform for experiment tracking and agent optimization. Every optimization run is tracked, visualized, and reproducible.
Try for free →
Afnio
The open-source Python SDK for building and auto-optimizing AI agent workflows. Define your harness, call optimizer.step(), and let Afnio do the rest.
118 elements in the periodic table. We're just getting started.
Examples
Built with Tellurio
Build self-improving browser agents
Browser agents that learn to solve new challenges from each new browser session.
Learn more →Align complex evaluation harness
Design multi-step evaluation pipelines, score agent outputs against ground truth, and auto-optimize to beat your accuracy targets.
Learn more →Optimize classifier accuracy
Auto-optimize prompts using textual gradients. Reach production-grade precision without manual prompt engineering.
Learn more →Ship your first self-improving agent in minutes
Free to start. No credit card required. Deploy optimized agents in minutes.