Enter a job title or keyword
Founding Reinforcement Learning (RL) Engineer
Job Location:
Monthly Salary:
Experience Required:
Posted:
24 October 2026 (11 hours ago)
Posted:
Application Deadline:
21 January 2027
Application Deadline:
Vacancies:
1 Vacancy
Vacancies:
Job Summary
Founding RL Engineer
San Francisco CA on-site Full-time
The build: Our client is hiring a Founding Reinforcement Learning Engineer to build the core RL systems from the ground up. Youll work directly with the founders own the full loop from environment design to training to evaluation and help shape the technical direction of the company.
What youll do
- Design and build RL environments reward functions and training pipelines.
- Train and fine-tune models with RL methods (PPO GRPO DPO RLHF/RLAIF and similar).
- Build evaluation frameworks to measure model and agent performance.
- Run experiments fast read results and decide what to try next.
- Scale training on GPU clusters and keep pipelines reliable.
- Turn research ideas into production systems.
- Help set engineering culture and hire the next engineers.
Requirements
What were looking for
- 2 years of hands-on experience in reinforcement learning or ML engineering.
- Strong Python skills and deep experience with PyTorch (or JAX).
- Hands-on work training models with RL (policy gradient methods reward modeling RLHF or similar).
- Experience with LLM post-training or agent training is a big plus.
- Comfortable with distributed training and GPU infrastructure.
- Builder mindset: you ship fast and are happy in a small early team.
- Degree in CS Math Physics or a related field (MS/PhD a plus).
Nice to have
- Publications or open-source work in RL.
- Experience at an AI lab or an early-stage AI startup.
- Experience with simulation or environment frameworks (Gymnasium Ray RLlib Isaac MuJoCo).
- Experience with training libraries like TRL verl or OpenRLHF.
Logistics: San Francisco CA on-site.
Required Skills:
Python PyTorch or JAX reward modeling RL training policy gradient methods PPO GRPO DPO RLHF/RLAIF environment design evaluation frameworks distributed training GPU infrastructure LLM post-training (plus) agent training (plus) Gymnasium/Ray RLlib/Isaac/MuJoCo (plus) TRL/verl/OpenRLHF (plus)
Required Education:
Bachelors degree from an accredited college or university is preferred but not required.
