Principal Software Engineer

Job Summary

A company is looking for a Principal Software Development Engineer specializing in LLM reinforcement learning.

Key Responsibilities

• Design, implement, and tune post-training methods on large-scale HPC clusters
• Develop high-throughput synthetic-data pipelines with verifiable results
• Collaborate with various teams to integrate metrics and publish improvements under permissive licenses

Required Qualifications

• Proven experience in a post-training environment, particularly with reward modeling and RL at scale
• Track record of engineering success or research contributions, including publications or open-source releases
• Solid foundation in statistics, optimization, and error analysis
• Familiarity with industrial research workflows and modern software engineering practices
• Post-graduate degree in a relevant field (ML, NLP, RL) is preferred