Staff Research Engineer

Role overview This staff-level position leads research engineering work on frontier AI systems, designing datasets, reinforcement learning environments, and benchmarks that improve advanced model capabilities across domains. The role spans foundational research through applied development, with ownership over technical direction and the autonomy to translate findings into production-grade improvements.

Responsibilities - Lead investigations into the capabilities, limitations, and training methods of frontier AI systems across domains such as software engineering, finance, retail, and developer tools. - Design rigorous experiments, prototypes, tooling, and evaluation frameworks for synthetic data generation, reinforcement learning, post-training, and model understanding. - Build and operate RL environments for coding agents, computer-use or browser-use agents, and general function-calling workflows. - Translate research findings into scalable product and platform improvements through close collaboration with research, engineering, product, and operations teams. - Contribute to the broader research community through publications, open-source work, workshops, conference participation, and mentorship. - Establish and uphold practices for experimental rigor, data quality, reproducibility, and interpretation.

Requirements - Substantial research background in machine learning, AI systems, or related fields, with demonstrated technical leadership. - Deep proficiency with modern AI and ML tools for training, testing, and evaluating models at scale. - Track record of designing experiments that yield evidence-based, reproducible conclusions. - Strong communication skills for both specialized researchers and cross-functional partners. - History of contributing to AI research through technical publications, open-source work, or industry equivalents.

Nice to have - Experience building RL environments or benchmarks for coding agents, browser-use agents, or enterprise function-calling systems. - Background in long-range, multi-step workflow design and evaluation. - Familiarity with agentic AI systems and enterprise deployment contexts.

Benefits and work setup - Compensation range of $250,000 to $400,000 OTE plus equity. - Opportunity to set technical direction on post-training, RL environments, and frontier benchmarks. - Potential to publish at leading AI conferences including ICLR, ICML, and NeurIPS. - High-autonomy environment with significant commercial impact and rapid iteration.