Description
ai& is hiring a worldwide Post-Training and Reinforcement Learning Engineer to own end-to-end post-training for internal custom models and enterprise customers. The role combines research and engineering across reinforcement learning, SFT, preference alignment, reward models, synthetic data, continual learning, evaluation, and production pipelines. Responsibilities include running and optimizing RL experiments, building training infrastructure, translating customer requirements into post-training workflows, designing evaluations, and contributing to research. The role requires hands-on experience with language-model post-training, Python and PyTorch, training frameworks such as DeepSpeed or FSDP, data-quality workflows, and customer-facing technical ownership.
