About

About me and my background

I’m an ML engineer who builds production systems that actually work. I care deeply about the intersection of cutting-edge research and practical infrastructure.

What I Do

I build systems that bridge the gap between cutting-edge ML research and production reality. My work centers on making complex models actually work at scale not just in notebooks, but in the messy, unpredictable world of real users and real data.

On the infrastructure side, I focus on automation, observability, and reliability. I believe in boring technology that works over exciting technology that doesn’t.

When I’m not wrangling machine learning pipelines or debugging gradient explosions, I’m probably reading papers on transformers, diffusion models or thinking about how to make ML systems more reproducible.

My principles

Models are only as good as their deployment. The gap between a working notebook and a production system is where the real engineering happens.

  • First principles thinking over cargo-culted best practices
  • Boring, reliable infrastructure over exciting but fragile systems
  • Reproducible experiments over one-off successes

Technology

ML & AI

PyTorch JAX Transformers Diffusion Models MLflow Hugging Face

Infrastructure

Kubernetes Docker AWS AWS SageMaker Terraform Gitlab CI/CD

Languages

Python Go C++ Bash SQL

Misc

GPU optimization Model quantization Distributed training

Connect

I’m always interested in discussing ML systems, infrastructure challenges, or interesting technical problems.