Research Engineer, Evals

About the Role

Build evaluation systems that measure the quality of large-scale web data extraction. You will define what good output means, develop the systems and datasets to assess it, and connect evaluation results to model and product decisions.

What You'll Do

• Design rigorous metrics for extraction quality across diverse websites, formats, and edge cases.

• Build scalable pipelines and evaluation harnesses to measure output quality.

• Generate and curate trustworthy datasets with reference extractions and ground-truth labels.

• Use evaluation results to guide model and product decisions, closing the feedback loop from measurement to improvement.

• Make evaluation findings clear and actionable for the broader team.

What We're Looking For

• At least 2 years of relevant evaluation experience, with 4 to 10 or more years of overall experience.

• Experience designing and building evaluation systems for LLM-based structured extraction from messy, real-world web pages or documents, including field-level accuracy measurement at scale.

• Experience building or evaluating large-scale web or document scraping and extraction systems, and creating ground-truth datasets.

• Strong Python skills and experience with machine learning, LLMs, data pipelines, evaluation frameworks, web scraping, and NLP.

• Statistical rigor in metric design and interpretation, including sampling, estimator bias and variance, significance, error bounds, and judge-to-human agreement.

• A combination of research methodology and hands-on engineering, with examples of evaluations informing real model or product decisions.

Compensation & Benefits

Base salary range is $250,000 to $290,000 USD annually. Visa sponsorship is available.

Location

Hybrid, with a primary location in San Francisco, California. The role requires working on-site five days per week in San Francisco or Toronto, with relocation before starting if needed.