I'm an open-source engineer building infrastructure for reliable and language-agnostic AI agents.
I'm the creator of Inference Gateway, an open-source ecosystem for connecting, building, and operating AI agents across models, providers, programming languages, and environments.
Previously, I spent nearly two years at Tesla working on global engineering projects. Today, I'm focused on building open-source systems that make agent orchestration more reliable, observable, and accessible.
-
Inference Gateway
A unified gateway for routing inference across multiple AI providers and self-hosted models. -
Infer
A Git-first coding agent designed for delegation, automation, and real software-engineering work. -
Infer Action
A GitHub-native workflow for turning issues into implemented and reviewable pull requests. -
Agent Definition Language
An open, language-agnostic specification for defining portable AI agents.
- Reliability matters more than impressive demos.
- Automation should remove work, not create more work to review.
- Deterministic systems should remain deterministic.
- Open standards are better than framework lock-in.
- The right abstraction is usually simpler than the first one.
Languages: Go, Rust, TypeScript, C# and C++
Infrastructure: Kubernetes, Google Cloud, AWS, GitHub Actions and GitOps
Focus: AI agents, distributed systems, observability, automation and
platform reliability






