About The LLM Data Company
The LLM Data Company (YC X25) provides post-training data and RL environments to foundation model labs and frontier applied AI companies. We have raised $3.6m from Tier 1 VCs and are growing 200%+ month-over-month.
Responsibilities
Design and implement scalable RL recipes for post-training task-specific models
Develop modular environments, reward functions, and evaluator scaffolds for internal and customer-facing tasks
Drive research at the intersection of scalable infra and modern RL frameworks to enable RL-as-a-service
Drive foundational research to publish open source environments and training data
Build data generation and curation pipelines to support frontier post-training
Collaborate with product teams to deliver a user friendly interface for non-technical users to generate data
Qualifications
Master or PhD in Computer Science or related field
Comfort with core tooling (verl, PyTorch, etc.)
Familiarity with modern post-training techniques (GRPO, etc.)
Experience with evaluations and reward engineering
Published in top journals (ICLR, NeurIPS, ICML, etc.)
Why you should join
Cutting-edge research: Work on unpublished, novel training environments
Direct lab exposure: Projects that labs actually use and validate in production
High autonomy: Wide design space to propose and run experiments with minimal oversight
Early team member: Join as one of the first 10 people with significant equity upside
Top Skills
What We Do
Post-training data research
