Prime Intellect
Research Engineer - Pre-training Data
$350K
San Francisco +1 · Full-time · Posted Oct 16, 2026
- Location
- San Francisco
Remote - Visa sponsorship
- Visa sponsorship and relocation assistance for international candidates.
About the job
Build and evaluate scalable training-data pipelines and corpora for Prime Intellect’s open models, working with researchers to improve pre-training and midtraining data.
What you'll do
- Find and collect valuable web, code, and technical training data; process it into training-ready corpora.
- Develop data quality, coverage, and contamination checks, and make large-scale processing efficient and reproducible.
- Run data-mixture experiments and use model evaluations to assess data for reasoning, coding, and long-context capabilities.
What they ask for
- Strong Python programming skills and a record of building reliable data-processing or distributed systems.
- Experience with large datasets and practical tradeoffs in storage, compute, throughput, and reproducibility.
- Ability to diagnose data-quality issues and translate research goals into measurable checks and experiments.
- Comfort owning evolving technical problems end to end and working closely with researchers.
Nice to have
- Experience preparing datasets for language-model pre-training, continued pre-training, or midtraining.
- Experience with distributed processing frameworks such as Spark or Ray.
Pay
$150-350k (cash compensation, including equity incentives; currency not stated)
Tools
Python, Spark, Ray
AI focus
Prepare and evaluate training data for open-model pre-training and midtraining, including data for reasoning, coding, and long-context learning; test synthetic data where useful.
Model-written summary · the original posting is authoritative