This is a fully remote position, open to applicants in United States, +3 more locations.

📋 Description

• Transform workflows into actionable tasks, environments, evaluations, and reward signals.

• Conduct post-training experiments against actual baselines and implement successful outcomes.

• Assess whether improvements originate from the model, data, harness, inference, or infrastructure.

• Develop significant open-source reinforcement learning (RL), environment, evaluation, and agent systems.

• Influence researcher-facing features in the training platform through engineering and design collaboration.

• Determine which abstractions should evolve into stable interfaces.

• Generate APIs, examples, reference implementations, and technical documentation.

• Collaborate with research users and external contributors to enhance interfaces, code, and research outputs.

• Assist new researchers in advancing from conceptual environment ideas to credible evaluations and training executions.

• Establish contributor pathways for external researchers to deliver environments, harnesses, algorithms, and integrations.

⛳️ Requirements

• Relevant experience in managing open-source machine learning (ML) or developer-tool systems, research engineering, technical product roles, or communities centered around code and reproducible research.

• Practical experience in constructing and troubleshooting model-training, post-training, agent, evaluation, or closely related experimental processes.

• Strong experimental judgment, including the ability to set meaningful baselines, perform meticulous trace inspections, attribute failures clearly, and adapt direction based on changing evidence.

• Proficient in Python programming.

• Capability to create artifacts that other researchers and engineers can execute, examine, and build upon.

• Comfort in tackling ambiguous problems while keeping partners and team members informed.

• Demonstrated exceptional work through publications, open-source systems, benchmarks, or delivered products.