Research Scientist, Frontier AI Data & Model Training — SF

Idlexperts; San Francisco (CA); On-site; USD 120,000 - 190,000; Full time

Benefits offered by this job

Free meals in office

Healthcare

401(k)

15 days PTO per year

Relocation assistance

Meaningful equity

Job summary

idler, a frontier data research lab in San Francisco, is hiring a Research Scientist to measure and improve how models learn from our tasks. You’ll validate data quality, identify where improvements are needed, and create new datasets in collaboration with frontier researchers.

To succeed, you’ll need extensive experience doing this work in production at a frontier lab. In this in-person role, you’ll design post-training experiments, build scalable data-processing systems, and contribute to our

Qualifications

• 1+ years of experience doing RL in production at a frontier lab.
• Track record of post-training an LLM end to end.
• Desire to drive the research roadmap and implementation on a fast-moving team.
• Deep curiosity about how machines learn from data and how to extract the maximum learning signal from our tasks.

Responsibilities

• Work with customers — researchers at frontier labs — to design novel post-training recipes and data quality measurement techniques.
• Design and run our in-house post-training stack to measure model lift on our tasks.
• Develop new data products based on datasets and experts available to us.
• Work with procurement to ensure external data we acquire is suitable for refinement.

Skills

RL in production

Post-training LLM

Data quality

Data products

Job description

idler, a frontier data research lab in San Francisco, is hiring a Research Scientist to measure and improve how models learn from our tasks. You’ll validate data quality, identify where improvements are needed, and create new datasets in collaboration with frontier researchers.

To succeed, you’ll need extensive experience doing this work in production at a frontier lab. In this in-person role, you’ll design post-training experiments, build scalable data-processing systems, and contribute to our