back

i'm currently a software engineer at commandcode.ai, handling inference infrastructure and devops since 25 june.

i am not open to work at the moment, but this page stays here as a snapshot of the kind of work i usually do.

llm api gateways, routing layers, fallbacks, provider health, and usage billing

backend systems in go, python, typescript, rust, and low-level linux tooling

high-throughput apis, queues, postgres, redis, observability, and production debugging

docker, kubernetes, ci/cd, aws, grafana, prometheus, and deploy pipelines

you have an infra-heavy product that needs someone who can ship without hand-holding

your backend is getting slow, flaky, expensive, or hard to operate

you need ai infra that works in production instead of just looking good in a demo

you want a builder who can go from idea to api to deployment to metrics

ex-cto at routing.run; left after it got acquired by an individual

built routing.run, an llm gateway with routing, fallback, auth, billing, and health checks

built rust infra at ultrabalancer reaching 500k+ rps

worked on go backend, swift macos, vllm/pytorch inference, and observability at megallm

email resume