Member of Technical Staff — RL Research

About the Company

Nuance Labs is building photorealistic, real-time AI avatars with emotional intelligence: a full-duplex audiovisual system that can listen, speak, react, interrupt, and respond like a real person.

Responsibilities

• Build Nuance’s RL/post-training stack from 0→1 including rollout generation, policy optimization, and reward modeling
• Develop and scale post-training methods such as PPO, GRPO, DPO, and RLHF/RLAIF
• Design system abstractions for trainers, rollout workers, and data queues
• Build evaluation and feedback loops for omni behavior like turn-taking and emotional response
• Optimize the end-to-end post-training loop for throughput, latency, and GPU utilization

Requirements

• PhD in ML, RL, or a related field (completed or in final stages)
• Solid understanding of RL/post-training methods and preference optimization
• Ability to reason about model behavior like reward hacking and distribution shift
• Exposure to RL/post-training pipelines or frameworks like verl, ms-swift, or OpenRLHF
• Strong software engineering fundamentals

Preferred Qualifications

• Hands-on experience with multimodal post-training for audio-video-language models
• Experience with PPO, GRPO, DPO, or synthetic data generation
• Prior 0→1 experience building RL pipelines or agent training systems
• Experience with distributed pretraining or inference serving

Benefits

• HSA plan with ~$2,000 annual company contribution
• 15 days PTO plus public holidays and a year-end office closure
• Daily lunch, drinks, and snacks
• Commuter benefits
• Visa sponsorship available (O-1, H-1B, green card)