I enjoy building things from the ground up, and I’m currently focused on observability, reliability, and resource efficient on applied AI/ML systems.
Currently, I put my spare time to built/maintain some tools which I use it as a part of my workflow:
- Orkes: A lightweight orchestration framework for LLM agents, no abstractions you don’t need, just clean logic that works.
- Orkes Watcher: observability platform for Orkes, graph-based agents. It just answers, for any run, exactly what happened.
In addition, I love exploring emerging technologies in the AI/ML space and integrating them into decent MVPs. I update my experimental repository from time to time with my latest findings [link].
Always down to collaborate with curious minds. If you have an idea or just want to talk tech, hit me up!
