AI Benchmark & Datasets Engineer / Researcher

Pathway builds the first post-transformer frontier model that solves AI's fundamental memory problem, enabling true continuous learning, infinite context reasoning, and real-time adaptation. The company combines this foundational model with the fastest data processing engine on the market and is trusted by organizations such as NATO, La Poste, and Formula 1 racing teams. Pathway is led by AI pioneers including CTO Jan Chorowski and CSO Adrian Kosowski, and is headquartered in Palo Alto, California.

• Published at least one paper at NeurIPS, ICLR, or ICML as lead author or with significant conceptual and code contributions, or
• Significantly contributed to an LLM training effort which became newsworthy, preferably using multiple GPUs, or
• Spent at least 6 months working in a leading Machine Learning research center (e.g., Google Brain, Deepmind, Apple, Meta, Anthropic, Nvidia, MILA), or
• Was an ICPC World Finalist, or an IOI, IMO, or IPhO medalist in High School.
• Experience with ML/LLM evaluation, data science, or technical product roles, ideally around benchmarks or experimentation.
• Comfortable reading papers, leaderboards, and Github repos, and turning them into clear, repeatable benchmark specifications.
• Ability to communicate effectively with engineers and customers, translating between technical detail and business value.
• Care about high-quality data, reproducible experiments, and crisp documentation.
• Respectful of others.
• Fluent in English.

Bonus Points:

• Published or open-sourced work on LLM evaluation, benchmarking, or data quality.
• Experience designing custom benchmarks or evaluation protocols for novel model capabilities.

• Proactively identify, prioritize, and curate relevant public and client-driven benchmarks across target use cases and markets.
• Evaluate candidate benchmarks for clarity, data quality, evaluation methodology, and fit with the model roadmap.
• Run benchmarks with baseline models to validate setup, uncover edge cases, and de-risk R&D runs.
• Hand off "benchmark-ready" packages to R&D including specs, data, evaluation scripts, expected metrics, and constraints.
• Maintain shared vocabulary and documentation around benchmarks, datasets, and evaluation formats usable by GTM and R&D.
• Track and organize benchmark results, model leaderboards, and define "what good looks like" for different customers and scenarios.
• Contribute to demos and public-facing proof points based on benchmark outcomes.

Candidates based anywhere in the EU, UK, United States, and Canada will be considered.