ML Infra / Orchestration Infrastructure Team

• Managed and maintained a distributed infrastructure integrating Airflow, Temporal, Spark, and Kubernetes for scalable workflow orchestration with over a 1,000,000 actions a month.

• Developed a high-performance gRPC server in Kotlin, enabling seamless communication with Kubernetes pods managed by Airflow.

• Built a Slack bot leveraging semantic search and LLMs to efficiently resolve triage questions.

• Orchestrated the deployment of the entire team's infrastructure and environments in the UK, ensuring flawless operational continuity.

• Instrumented tooling for worker profiling inefficiencies: CPU throttling from 70% peak to ~10%.

• Installed KEDA autoscaler, adjusted health checks: worker startup reduced by ~50%.

• Led team of 5 engineers to create training data tooling and ~200 LLM evals for Devin AI at Cognition Labs.

• Established a QA rotation and procedure for engineers

*Please do not contact me for “networking calls” to learn more about Devin/Cognition. Thanks!

Boutique Software Contracting Firm

• Lead a team of 4

• Secured over 5 contracts for custom software projects

• Achieved a peak quarterly revenue of ~$20,000