Periodic Labs · 2 months ago
Research Engineer - Midtraining
Periodic Labs is an AI and physical sciences lab focused on making novel scientific discoveries through advanced modeling. The role involves training frontier models, developing synthetic data generation methods, and collaborating with researchers to create evaluations for scientific data curation.
Artificial Intelligence (AI)Foundational AIGenerative AIMachine Learning
Responsibilities
You will train frontier models to be highly knowledgeable scientific experts that serve as the foundation for reinforcement learning
You will develop methods for synthetic data generation, distillation, and continual learning at scale
You will work closely with RL researchers, physicists, and chemists to create evals that guide scientific data curation
You will collaborate with supercompute engineers to scale compute-efficient LLM training to thousands of GPUs
You will build high-performance tools for yourself to investigate how data shapes intelligence
Qualification
Required
Experience with training LLMs on curated mixes of trillions of tokens
Experience with calculating scaling laws and compute-optimal hyperparameters
Experience with generating billions of tokens of high-quality synthetic data
Experience with building evals that correlate with downstream task performance
Company
Periodic Labs
Periodic Labs develops artificial intelligence systems that simulate and predict the properties of materials using machine learning.
H1B Sponsorship
Periodic Labs has a track record of offering H1B sponsorships. Please note that this does not
guarantee sponsorship for this specific role. Below presents additional info for your
reference. (Data Powered by US Department of Labor)
Distribution of Different Job Fields Receiving Sponsorship
Represents job field similar to this job
Trends of Total Sponsorships
2025 (2)
Funding
Current Stage
Early StageTotal Funding
$300M2025-09-30Seed· $300M
Recent News
Business Insider
2025-12-26
Inside HPC & AI News | High-Performance Computing & Artificial Intelligence
2025-12-20
Company data provided by crunchbase