attention head AI jobs Alerts Saved
AI jobs › Post-training & RL AI jobs

Post-training & RL AI jobs

Post-training takes a pretrained model and shapes how it behaves: following instructions, holding a conversation, reasoning through hard problems, writing code, using tools, and declining what it should decline. The main methods are supervised fine-tuning on example answers, preference methods such as RLHF and DPO, and reinforcement learning against rewards that can be checked, such as passing tests or reaching a correct answer.

At labs, research scientists and research engineers design these recipes, train reward models, build RL environments and the infrastructure that runs RL at scale, and work out why a model got worse at one thing while getting better at another. Outside labs, ML engineers and applied AI engineers fine-tune open models on their own data, often with parameter-efficient methods such as LoRA, to make a model cheaper or better at one task. Expect PyTorch, a feel for data quality, and careful evaluation.

We count roles whose postings mention post-training, RLHF or RLAIF, reinforcement learning, reward models, or fine-tuning. Fine-tuning is the broadest of these: it brings in many applied roles that adapt existing models rather than train new ones.

1,432 open AI roles at 614 employers, from their own job boards. Showing the newest 100.

Follow Post-training & RL102 new this week

New Post-training & RL roles by email, the morning after we find them, in one email with your other alerts. Narrow it, or keep every role.

How often
We email a link to confirm it.
Amazon · 83NVIDIA · 48Anthropic · 28Microsoft · 25Waymo · 24Mistral AI · 22OpenAI · 22Scale AI · 21Neura Robotics · 15Prime Intellect · 15Fireworks AI · 14Mercor · 12Lila Sciences · 11Liquid AI · 10Articul8 · 9

Get new Post-training & RL roles every Monday

A job alert: new Post-training & RL roles from employers' own job boards, every Monday. Unsubscribe in one click.

Related