NewsBreak
Machine Learning Engineer, LLM Post-Training
$150K–$230K
Mountain View, California, United States · Full-time · Posted Jun 10, 2026
About the job
Own the LLM post-training pipeline, with a primary focus on reinforcement learning, and deliver product-targeted model improvements.
What you'll do
- Lead continuous pre-training, supervised fine-tuning, and reinforcement learning of LLMs.
- Design and curate training data, and build evaluation and reward/verifier pipelines.
- Run large-scale distributed GPU training and partner with product and business teams to deliver model capabilities.
What they ask for
- Hands-on experience personally running CPT, SFT, and RL training, including practical RL experience.
- Ability to independently design ML data-preparation strategies for business needs.
- Experience training LLMs on mid-to-large GPU hardware and debugging distributed training at scale.
- Strong PyTorch fundamentals and understanding of tokenization, attention, chat templates, and alignment or agent-training failure modes.
Nice to have
- Experience designing reward models or rule-based verifiers, or training tool-use/agentic models.
- Publications or open-source contributions in LLM post-training or RL.
Pay
$150,000 — $230,000 USD (annual base salary)
Tools
PyTorch, Hugging Face TRL/Accelerate, DeepSpeed, FSDP, vLLM, RLHF, PPO, GRPO
AI focus
LLM post-training, especially reinforcement learning, including CPT, SFT, reward/evaluation pipelines, and tool-use or agentic training.
Model-written summary · the original posting is authoritative