Lead Research Engineer, Data Quality (AI)

TL;DR

Lead Research Engineer, Data Quality (AI/RL): Owning the measurement, improvement, and scaling of training data quality for frontier AI agents with an accent on QC systems, synthetic data validation, and evaluation frameworks. Focus on turning qualitative research insights into production systems and building internal research taste for agent training data utility.

Location: On-site only in San Francisco or Singapore. Support for relocation and visas is provided for strong candidates.

Company

hirify.global is building infrastructure for RL training data and evals for frontier AI agents, including a marketplace for frontier labs.

What you will do

• Lead hirify.global's data quality strategy, including building QC systems and defining quality standards.
• Develop scalable methods for validating synthetic data, such as trajectory auditing and failure-mode analysis.
• Collaborate with research engineers and domain experts to diagnose and resolve data quality issues.
• Translate research insights into production systems, internal tools, and validation pipelines.
• Mentor other research engineers to maintain high technical rigor and execution speed.

Requirements

• Advanced proficiency in Python, Docker, and Linux environments.
• Deep intuition for data quality (realism, diversity, and reliability of training tasks).
• Experience building QC systems, evals, benchmarks, or synthetic data pipelines.
• Ability to work across messy human and technical systems involving vendors and domain experts.
• Strong written communication skills for explaining methodology to technical audiences.

Nice to have

• Experience leading teams on ambiguous technical projects.
• Ability to convert subject-matter expert judgment into scalable review systems.
• Experience designing metrics and QA/QC processes from scratch.
• Early-stage startup experience.

Culture & Benefits

• Comprehensive medical, dental, and vision insurance (US).
• 401k and commuter benefits (US).
• On-site meals (lunch and dinner).
• Company-wide holiday break from Christmas Eve to New Year's Day.
• Unlimited access to AI tools (ChatGPT, Claude Code, Cursor, etc.).
• Equinox membership.

Hiring process

• Two technical interviews.
• A 2-3 day work trial.