SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Salary: USD 150,000 - 350,000 / annual
Prime Intellect is building an open superintelligence stack—infrastructure that frontier AI labs typically build internally, now available to every ambitious AI team. The platform unifies compute, environments, evaluations, secure sandboxes, high-performance training, and deployment for post-training at frontier scale, from SFT and RL to tool use, agent workflows, and continuously improving production models. The company is also developing open-source models trained end-to-end for long-horizon tasks like autonomous research.
As a Research Engineer focused on Pretraining Data, you will own the data strategy behind Prime Intellect's next generation of open models. You'll identify valuable data sources, transform them into training-ready corpora, and collaborate closely with researchers to validate what actually improves model performance. Your work will directly shape both broad pretraining and specialized data used for reasoning, coding, and long-context capabilities.
Key responsibilities include:
- Finding, evaluating, and collecting valuable training data from web, code, and technical sources
- Building scalable pipelines that convert raw data into training-ready corpora through extraction, cleaning, deduplication, and filtering
- Developing quality checks for data coverage and contamination, then using them to improve collection and processing
- Partnering with researchers on data mixtures and sampling strategies for pretraining and midtraining
- Preparing and evaluating data for long-context learning, reasoning, and coding, including testing synthetic data where applicable
- Running controlled data experiments and using model evaluations to decide which sources, mixtures, and processing choices merit scaling
- Ensuring large-scale data processing is efficient, reproducible, and recoverable, with clear source provenance and dataset versioning
Prime Intellect has raised $150M from Founders Fund, Radical Ventures, NVIDIA, and exceptional AI and infrastructure operators including Andrej Karpathy, Dwarkesh Patel, and leaders from Ramp, Perplexity, Harvey, Mercor, Zapier, Datadog, Cognition, OpenAI, and others.
REQUIREMENTS:
- Strong programming skills in Python and a proven track record building reliable data-processing or distributed systems
- Experience working with large datasets and managing practical tradeoffs of storage, compute, throughput, and reproducibility
- Ability to diagnose data-quality issues and translate research goals into measurable checks and experiments
- Comfort owning an evolving technical problem end-to-end and working closely with researchers to test assumptions and improve the system
Relevant experience may include: web crawling, document extraction, code-data collection, or large-scale corpus curation; deduplication, text classification, quality scoring, synthetic data, or training-data contamination analysis; data-mixture experiments, curriculum learning, or measuring how dataset changes affect model performance.
Nice to have: experience preparing datasets for language-model pretraining, continued pretraining, or midtraining; work with distributed processing frameworks such as Spark or Ray; experience with multilingual, mathematical, scientific, or repository-level code data; familiarity with tokenization, training data loaders, and the needs of large model-training runs.
