Senior Staff Engineer, RL Post-Training Infra

Embedding; San Francisco (CA); Vor Ort; USD 180.000 - 280.000; Vollzeit

Zusammenfassung

Moonlake is hiring a Member of Technical Staff to lead reinforcement learning infrastructure and model post-training in San Francisco. You will work with the research team to improve large vision-language and code-generating agents through fine-tuning, reinforcement learning, trajectory data, and scalable evaluation.

We are building AI infrastructure for interactive digital worlds and robotics; this role shapes the training systems that enable agents to plan, code, use tools, and operate in

Qualifikationen

• Real-world experience training large language, vision-language, multimodal, or code models.
• Strong experience in reinforcement learning, post-training, or large-scale fine-tuning.
• Experience building distributed training or high-throughput inference systems.
• Familiarity with supervised fine-tuning, preference optimization, reward modeling, and agentic RL.

Aufgaben

• Build RL and post-training pipelines for multimodal, vision-language, and code agents
• Develop infra for supervised fine-tuning, preference modeling, reward shaping, and RL
• Create systems to collect, filter, replay, and learn from agent trajectories
• Design rewards, verifiers, and evaluations for long-horizon tasks
• Improve agents' ability to plan, code, use tools, and recover from errors
• Scale distributed training and high-throughput rollout across multi-GPU clusters
• Improve training reliability, reproducibility, observability, and cost efficiency

Kenntnisse

Python

Reinforcement Learning

Distributed training

Research judgment

Tools

PyTorch

JAX

Jobbeschreibung

Moonlake is hiring a Member of Technical Staff to lead reinforcement learning infrastructure and model post-training in San Francisco. You will work with the research team to improve large vision-language and code-generating agents through fine-tuning, reinforcement learning, trajectory data, and scalable evaluation.

We are building AI infrastructure for interactive digital worlds and robotics; this role shapes the training systems that enable agents to plan, code, use tools, and operate in