Description
OpenAI is hiring an Agent Post-Training, Computer Use specialist to teach frontier models how to navigate browsers and desktops, use tools and applications, reason through complex workflows, collaborate with users and other agents, and complete long-horizon tasks. The role owns experiments, post-training stacks, evaluation environments, data pipelines, reward signals, diagnostics, and model-behavior analysis; partners with researchers, engineers, product, infrastructure, and safety teams; and helps decide which capabilities and fixes enter major model runs. Candidates should have strong technical fundamentals and hands-on experience with LLMs, reinforcement learning, post-training, evaluation, synthetic data, coding or tool-using agents, or production machine-learning systems.
