Members-Only
Recent Talks & Demos are for members only
You must be an AI Tinkerers active member to view these talks and demos.
Invariant OSS: Debugging AI Agents
A walkthrough of open-source invariant tools for debugging computer‑use agents like OpenAI’s Operator, showing practical steps to trace actions and understand model behavior.
A quick walkthrough of some Invariant OSS tools that we are working on and how to use them to debug Computer Use Agents like OpenAIs recently announced Operator.
Invariant provides Python guardrails intercepting LLM agent tool calls for security.
Visualizes, inspects, and analyzes AI agent traces using Docker Compose.
Related projects
Agents That Debug Humans
Amsterdam
Live coding how agents automatically detect faulty user configurations, rewrite rules, and verify fixes, providing production‑ready self‑healing for…
Calmo Demo
Amsterdam
A live demonstration of Calmo, an AI agent that analyzes production alerts, isolates root causes, and automates fixes,…
EidOS, an Agent Operating System
Seattle
Learn how to simplify LLM data products using a multi‑agent approach, then watch a live coding session building…
Agent Laboratory: Agentic AI Researcher
Prague
The session demonstrates an end‑to‑end autonomous workflow using specialized LLM agents to conduct literature reviews, design plans, run…
Why do agents keep making the same mistakes? 🤯 Let's fix it!
Munich
Learn how to build self‑improving agents that fetch today’s calendar, differentiate internal from customer meetings, and automatically generate…
Agents using Advanced RAG Capabilities
Dubai
Demonstrates a proof‑of‑concept using modern retrieval methods—late interaction, OCR‑free pipelines, reasoning retrievers, and hypothetical embeddings—to improve agentic systems…