Researcher, Connectors - Agent Post-Training

Train and evaluate agent models to connect LLMs to professional software and ship model improvements.

• •Train and evaluate agentic models to interface with professional software, build data, evals, and training pipelines, and ship improvements into products.
• •Key Responsibilities Design and run experiments improving agent behavior for complex software and plugins.
• •Own post-training stack improvements: RL, data pipelines, graders, reward signals, and evals.
• •Build evals/environments, convert failures into training data or product fixes.
• •Partner with product teams to translate product signals into model improvements.
• •Improve large-scale training machinery for reliability, observability, and production readiness.
• •Requirements Strong technical fundamentals across ML, software engineering, systems, or statistics.
• •Hands-on experience with LLMs, RL, RLHF/RLAIF, post-training, evals, or graders.
• •Experience building experiments, synthetic data, and evaluation loops for model behavior.
• •Ability to work cross-functionally and debug hard failures into concrete fixes.