I am a software engineer passionate about distributed systems, performance engineering, and infrastructure. I am currently building Orpheus, an open-source runtime for AI agents with queue-driven autoscaling.

I spent 2025 building Zoops, a GTM automation platform powered by AI agents. While building the product, developed infrastructure for agent execution: queue-based autoscaling, tier-based SLAs, and framework-agnostic orchestration. After running it in production for several months, I realized the infrastructure layer I'd built was solving a more fundamental problem than the application itself. I extracted it cleanly (the architecture made this possible in just a few hours), and that became Orpheus.

Before Zoops, I was one of the founding engineer at ToolJet, an open-source low-code platform. I joined when the team was small and contributed across the stack — building the core product and architetcure, performance optimization (frontend and backend), and platform scaling. I spent 3 years there, taking the product from early stage to enterprise scale.

My technical interests span the full stack: frontend performance, backend systems and orchestration engines, and distributed systems architecture. I've worked on DAGs, workflow engines, execution runtimes, and system design at scale.

I'm particularly interested in how systems behave under failure, how to scale on the right metrics (queue depth, not CPU), and how to build infrastructure that developers actually want to use.

I write about systems, performance, and infrastructure on Medium.