Research Engineer
Core
Design metrics, build pipelines, and generate datasets to rigorously evaluate the quality of Firecrawl's web-to-data conversion system.
Role type
Research Engineer (Evaluation Systems)
Builds
Evaluation pipelines and datasets for an AI data extraction product
Domain
AI Infrastructure / Web Data Extraction
Deliverable
production ML models
Required skills
metric design, pipeline engineering, dataset curation, feedback loop implementation, ambiguity resolution, rapid iteration
Preferred skills
building evaluation systems from scratch, defining methodology for ambiguous problems
Technologies
(none explicitly listed)
Responsibilities
Design metrics defining "good output" across millions of sites; Build pipelines to measure quality at scale; Generate and curate trustworthy datasets; Own the feedback loop from output quality to model/product decisions; Turn evaluation results into actionable team insights
Seniority
Mid-Senior, hands-on IC