Software Engineer
Role Overview
We are hiring for one of our clients, seeking a Engineering & Software Domain Expert to work on a full-time basis. Your engineering expertise will fuel the development of the most advanced AI models by reviewing engineering knowledge work tasks, writing instruction specs and golden solutions, and building benchmarks to measure model improvements. This role requires a practicing specialist who can define what "correct" looks like in real-world engineering and software development work.
Key Responsibilities
- Vet the quality of engineering knowledge work tasks and model outputs, identifying missing behaviors, thin reasoning, code errors, unhandled edge cases, and answers that would not pass a real code review.
- Write high-quality instruction specifications and produce golden solutions to engineering problems.
- Design challenging engineering tasks and evaluation sets to assess model performance.
- Collaborate with research and program management teams to improve how frontier AI models reason about engineering and software development.
- Build engineering-specific skills and tools alongside research teams using client-issued accounts and equipment.
Required Skills & Qualifications
- Experience in software engineering with expertise in algorithms, data structures, and system design.
- Ability to review and evaluate code quality, identify edge cases, and assess model outputs against real-world engineering standards.
- Proficiency in writing clear, precise instruction specifications and technical documentation.
- Experience designing benchmarks and evaluation frameworks for software development tasks.
- Strong problem-solving skills with a focus on practical engineering challenges.
More About the Opportunity
This role offers a unique opportunity to work with a global leader in the Technology, Information and Internet industry, contributing to the development of cutting-edge AI systems. You will be placed at a leading AI lab as part of their extended workforce, working directly with their research teams.