Open to opportunities

Rahul Garg

@rahulgarg

Message

I build agentic AI training environments and distributed LLM inference systems.

India

Message

What I'm looking for

I'm looking to build agentic AI training environments, robust evaluation systems, and distributed LLM inference infrastructure where rigorous engineering and measurable model reliability matter.

At Bespoke Labs, I build and QC-review Dockerized RL environments, agentic DevOps tasks, verifiable graders, and evaluation rubrics for frontier-model post-training. I was promoted from RL environment developer to quality reviewer, auditing verifier rigor, reward-hacking resistance, and difficulty calibration across work produced by a 100+ engineer pool.

Previously at Tessell, I built a cloud-native DBaaS platform on Google Cloud from 0 to 1, redesigned its control plane to reduce infrastructure management overhead by 90%, and improved workflow execution times by 40%. I also built AWS operations tooling for SRE teams, high-performance Go and Java microservices, and distributed LLM inference systems including InferLab.

Experience

Work history, roles, and key accomplishments

Current

Senior Software Engineer

Current

Bespoke Labs

Dec 2025 - Present (8 months)

Promoted from RL environment developer to quality reviewer, auditing agentic tasks, RL environments, and rubrics produced by a 100+ engineer pool. Designed agentic DevOps tasks, RL environments, and evaluation rubrics for frontier-model post-training, and built Dockerized environments with hidden test-based verifiers.