Research Engineer RL Frontiers
Etched is an AI-hardware company building rack-scale frontier inference clusters.
Funding history
About Etched
Etched co-designs chips, racks, software, and manufacturing systems for efficient inference of frontier AI models, targeting throughput, latency, cost, and power efficiency across prefill and decode workloads.
Skills
Candidate Availability
Required and preferred rules are kept separate and reflect the wording in the original posting.
About the Role
You will own the end-to-end post-training stack for frontier models. You will build reinforcement-learning environments with ASIC, kernel, and supercomputing engineers; create custom data mixes; and develop benchmarks that improve real model productivity while preventing reward hacking. You will use proprietary simulators and environments to post-train models for demanding engineering tasks.
Requirements
- Hands-on post-training experience, including SFT and RL
- Experience shipping models that people used
- Experience improving a model on a task it was not built for
- PyTorch and distributed-training engineering skills
- Understanding of the systems side of reinforcement learning
- Ability to make decisions under compute and data constraints
Responsibilities
- Own the end-to-end post-training stack
- Build reinforcement-learning environments with ASIC, kernel, and supercomputing engineers
- Create custom data mixes for relevant tasks
- Create benchmarks that improve model productivity and avoid reward hacking
- Train and post-train models using proprietary simulators and environments
Benefits
- Medical, dental, and vision packages with generous premium coverage
- $500 monthly credit for waiving medical benefits
- $2,500 monthly housing subsidy for employees living within walking distance of the office
- Relocation support for moves to San Jose
- Wellness benefits for fitness and mental health
- Daily office lunch and dinner
- Unlimited compute budget subject to ROI justification