Machine Learning Engineer, LLM Post-Training

Job Summary

Required Qualifications

• Hands-on LLM post-training experience (CPT, SFT, RL training)
• Demonstrated practical RL experience (RLHF / PPO / GRPO / DPO or similar)
• Strong data engineering for ML (data-preparation plans, labeling strategies, synthetic data generation)
• Proven large-scale GPU training ability on mid-to-large hardware with distributed-training experience
• Strong PyTorch fundamentals; familiarity with Hugging Face TRL/Accelerate, DeepSpeed or FSDP; experience with inference engines like vLLM
• Understanding of tokenization, attention, and common failure modes in alignment/agent training
• Ability to work across research and product teams and deliver training plans for specific business needs
• Ability to design, implement, and maintain evaluation and reward/verifier pipelines

Desired Qualifications

• Hands-on experience with large language model post-training (CPT, SFT, RL)
• Experience with RL methods (RLHF, PPO, GRPO, DPO or similar) in production
• Strong data engineering for ML (data sourcing, cleaning, labeling, synthetic data generation)
• Experience training LLMs on mid-to-large GPU clusters with distributed training
• Proficiency with PyTorch and ecosystems like Hugging Face TRL/Accelerate, DeepSpeed or FSDP
• Understanding of tokenization, attention mechanisms, and alignment/agent training
• Strong cross-functional communication to translate business requirements into training plans
• Ability to ship production-ready code and maintain evaluation/reward pipelines