Pre-Training Data Research Engineer (AI)

Location: United States; Hybrid in San Francisco

Company

hirify.global, spun out of MIT CSAIL, builds general-purpose AI systems optimized for low latency, low memory usage, privacy, and reliability across data-center accelerators and on-device hardware.

What you will do

• Source web-scale and public datasets and build pipelines for filtering, deduplication, and decontamination.
• Design data mixtures and curricula for pre-training and midtraining, deciding what enters each training run.
• Build synthetic data generation systems with LLMs, structured prompting, and distillation.
• Run proxy experiments, ablations, baselines, scaling ladders, and held-out-loss evaluations to measure the impact of data changes.
• Build trusted evaluation sets and maintain data lineage so training runs remain traceable and reproducible.

Requirements

• Deep, hands-on experience with web-scale data sourcing and curation, synthetic data generation, data mixing and curriculum design, deduplication, and decontamination.
• Ability to reason from data decisions to training outcomes, including interactions with model size, architecture, and training dynamics.
• Experience designing experiments to test data changes, including baselines, scaling ladders, small-scale proxies, and held-out loss, with clear interpretation of results.
• Experience building and owning production data pipelines, not only research prototypes.
• Solid ML fundamentals and experience training, evaluating, and iterating on models in Python; PyTorch preferred.
• At least 3 years of relevant experience with an M.S., 1+ year with a Ph.D., or 5+ years with a B.S.

Nice to have

• Experience with distillation, long-context data, or architecture-specific data strategy.
• Experience designing evaluation frameworks and benchmarks for pretrained models.
• Data experience spanning both pre-training and post-training.
• Experience with small, efficient models or on-device deployment.
• Startup experience or a track record of connecting model work to product needs.

Culture & Benefits

• Small, research-oriented team where pre-training data decisions go directly into shipped Liquid Foundation Models.
• Competitive base salary plus equity; final offers reflect skills, experience, and location.
• Unlimited PTO and company-wide Refill Days with shared long weekends.
• Comprehensive health coverage, varying by location.
• Paid leave for pregnancy, parenting, medical needs, and family care.