You'll own human data projects from start to finish. A project might be 800 scanned medical documents reviewed by five annotators over three weeks, or thousands of physician ratings on model output for a research team. You scope it with the team that needs it, design and test the task, staff it, keep it on schedule and on budget, catch quality problems before they spread, and deliver results the customer can trust. This role is built for someone early in their career who wants to own outcomes quickly, and who is as comfortable in a spreadsheet as in a room with researchers.

Own human data projects end to end: scope them with the requesting team, set timelines and budgets, and deliver on them.

Design tasks and instructions with the requesting team, and pilot them on a small batch before anything scales.

Staff projects across our in-house annotators and partner vendors, and manage those vendor relationships day to day.

Run calibration before production. Every annotator passes a known-answer test before their work counts, and anyone who doesn't gets feedback and a retry.

Track throughput, cost and quality by annotator and vendor, and report progress to leadership every week.

Catch problems early (an annotator drifting, a vendor slipping, an instruction that isn't working) and fix them before they reach the customer.

Handle sensitive data, including protected health information, under strict security procedures.

Travel on short notice when a project needs someone on site.

Make every project easier to run than the last one.

What Success Looks Like

30 days: learn the machine. You've shadowed a live project, taken over its roster and returns intake, and met each of our annotation partners.

Human Data Project Lead

Responsibilities

Qualification

Required

Preferred

Company

Protege

More roles at Protege