Research Engineer, RL Frontiers

Sourced from jobs.oceans.ventures

Etched is hiring a Research Engineer for its RL Frontiers team to train proprietary frontier models that automate kernel development, ASIC engineering, production pipelines, and hardware operating systems. You will own the post-training stack end to end, building RL environments and custom data mixes to push models to SOTA on hard engineering tasks. The role is fully in-person in San Jose and blends research with systems engineering.

• Own the post-training stack end to end, including SFT and RL.
• Build RL environments with ASIC, kernel, and supercomputing engineers, rewarded by simulation, formal tools, tests, and hardware measurements.
• Create custom data mixes that unlock frontier performance on relevant tasks.
• Create benchmarks that drive real productivity and avoid reward hacking.

• Hands-on post-training experience including SFT and RL (e.g., PPO, GRPO), with shipped models in use.
• Track record of taking a model from bad to good on a task it wasn't built for, ideally with scarce data.
• Strong PyTorch and distributed-training engineering skills.
• Understanding of the systems side of RL to make the loop faster and cheaper.
• Entrepreneurial judgment under compute and data constraints.

• Experience with RL or post-training frameworks such as verl, OpenRLHF, TRL, or prime-rl.
• Experience with inference engines such as vLLM or SGLang.
• Agentic, multi-turn RL with verifiable rewards.
• Distillation and training small, specialized models for production.
• ASIC design or verification, accelerator kernels, or large-scale compute infrastructure.
• Former founder experience.