Member of Technical Staff, Research Engineering
Job Description:
About the Role
Join an AI research engineering team working at the frontier of reinforcement learning. You will turn experimental ideas into scalable environments, training pipelines, and evaluations that help advance modern AI models.
What You'll Do
- Architect self-contained RL environments for complex tasks, including reward functions, verifiers, and evaluation logic.
- Design and scale episode pipelines and multi-component training processes for reproducible experimentation.
- Build automated synthetic data generation systems to accelerate training while maintaining data quality.
- Develop AI-driven evaluation and quality assurance systems for automated grading, validation, and feedback.
- Fine-tune and optimize open-source RL models using generated datasets and custom training strategies.
- Create benchmarks to measure model capability, robustness, and data quality across tasks.
- Contribute to the release and analysis of evaluations on internal and external benchmark platforms.
What We're Looking For
- Deep reinforcement learning experience, including environment design and training dynamics.
- A strong track record building and scaling RL systems, pipelines, or experimentation frameworks.
- Proficiency in automation and synthetic data generation pipelines.
- Familiarity with automated evaluation, model validation, and quality assurance workflows.
- Experience fine-tuning and evaluating open-source machine learning models.
- Experience publishing benchmarks, evaluations, or research artifacts is valuable.
- Familiarity with evaluation frameworks and scalable infrastructure for large-scale RL experimentation is valuable.
- Strong technical writing and communication skills, with comfort working in fast-paced, collaborative research settings.
Compensation & Benefits
Base salary is $200,000 to $300,000 USD annually. The package includes equity, and performance-based bonuses may also be available. Visa sponsorship is not available.
Location
Fully remote.