Projects

Selected Projects

Production AI systems, evaluation tooling, and automation experiments I have built or am actively developing.

Hermes AI Agent

A self-hosted production AI assistant with Telegram and Slack integrations, model routing, and full infrastructure control.

AI Agent Telegram Bot Slack Bot Model Routing
View project →

LLM Eval Harness

An open-source evaluation framework for testing LLM agents, prompt regression, and multi-step task reliability.

Open Source LLM Evaluation TypeScript Testing Framework
View project →