Researcher, Connectors - Agent Post-Training

OpenAI•San Francisco, CA

8d•$250,000 - $380,000

About The Position

The Agent Post-Training team at OpenAI is responsible for training the models behind frontier agents, including those used in Codex, ChatGPT, and the API. These agents are designed to be persistent, proactive, and capable of operating computers, collaborating with humans and other agents, and expanding human potential. The team defines future agent capabilities, develops the training signals for these abilities, and conducts experiments to realize them. Their work encompasses areas such as coding, tool use, computer operation, multi-agent coordination, long-horizon execution, factuality, instruction following, calibrated reasoning, and taste. This team is central to developing new model capabilities, building the data, environments, graders, training methods, and feedback loops that shape OpenAI's next generation of agents, and then integrating these capabilities into major training runs and final products. In the role of Agent Post-Training, Connectors, the focus is on teaching models to interface with professional software using code. This involves training agents to utilize code, APIs, tools, and structured integrations to operate across applications like Slack, Google Workspace, GitHub, Notion, Linear, Salesforce, and other essential work systems. The goal is to enable models to perform actions within a user's digital context, such as retrieving information, updating systems, coordinating tasks, generating content, and executing multi-step workflows using existing tools. The role aims to leverage the world's leading productivity and enterprise software to create a powerful action surface for agents. Collaboration with researchers, engineers, product teams, infrastructure teams, and safety/alignment partners is crucial for deciding on model run content, evaluating success, and shipping improvements to user-facing products. This is a high-agency position for individuals who want their work to directly influence frontier models.

Requirements

Strong technical fundamentals in machine learning, software engineering, systems, statistics, or a related field, and can learn quickly across the parts you have not worked in before.
Hands-on experience with LLMs, RL, RLHF/RLAIF, post-training, evals, graders, synthetic data, model training, coding agents, tool-using agents, or production ML systems.
Excited by open-ended problems where the path is unclear, the signal is noisy, and the right answer requires both research taste and engineering execution.
Care about product impact and model behavior, not just benchmark movement. You have opinions about what makes an agent useful, reliable, honest, tasteful, and easy to work with.
Can move from a vague behavioral problem to a concrete experiment: define the hypothesis, build the pipeline, run the model, analyze the result, and decide what to do next.
Comfortable working across research, product, infrastructure, data, evals, and safety boundaries, and can communicate clearly with each group.
Like building load-bearing systems and processes when that is what the team needs, even if the work is not glamorous.
Want to train and ship the models that make agents genuinely useful for developers, enterprises, researchers, and everyday users.

Responsibilities

Design and run experiments that improve agentic model behavior for complex software and plugins.
Own end-to-end improvements to the post-training stack, including RL, data pipelines, graders, reward signals, evals, diagnostics, and model-behavior analysis.
Build evals and environments that expose the next set of model failures, then turn those failures into training data, product fixes, or new research directions.
Partner with Codex and ChatGPT product teams to understand what users need and translate product signal into model improvements.
Work on early-training and alignment interventions, including data mixtures, objectives, synthetic data, and eval loops that shape downstream agent behavior.
Help decide which integrations, capabilities, and fixes are ready for inclusion in major model runs.
Improve the machinery for large-scale training and launch: experiment velocity, reliability, observability, reproducibility, cost, latency, and production readiness.
Take on cross-functional projects that touch model training, product infrastructure, and the production agent harness, such as multi-agent systems or training directly against production-like environments.
Debug hard failures in shipped or near-shipped models and turn messy qualitative behavior into concrete hypotheses, experiments, and fixes.