Post-Training Research Engineer

Post-Training Research Engineer

NYC, hybrid

We're partnered with a highly technical, research-driven quantitative firm that's been investing heavily in next-generation AI capabilities and looking to add a Post-Training Research Engineer to a growing team at the intersection of machine learning research and large-scale systems.

In this role, you'll help advance the performance and behavior of state-of-the-art models through post-training techniques, large-scale experimentation, and reinforcement learning. You'll work alongside researchers and engineers tackling challenging problems in alignment, reasoning, evaluation, and model optimization.

What You'll Work On:

• Post-training large language and foundation models
• RLHF, DPO, preference optimization, and reward modeling
• Model alignment and reasoning capabilities
• Automated evaluation frameworks and benchmarking
• Fine-tuning and large-scale experimentation
• Distributed training and GPU-intensive workflows
• Research that moves beyond papers and into production systems

Required:

• BS or higher from a top-tier academic institution in Machine Learning, AI, Computer Science, Mathematics, Statistics, or a related field
• Experience with post-training, RL / RLHF, DPO, reward modeling, alignment, and model evaluation
• Strong hands-on experience with Python and PyTorch
• Familiarity with distributed training frameworks and large-scale compute environments
• Track record of publishing research and/or deploying advanced ML systems into production
• Ability to bridge research and engineering

Target Background:

• Candidates from frontier AI labs, leading research organizations, and teams building large-scale foundation models are especially relevant.

This is an opportunity to work on some of the most challenging problems in modern AI, with significant resources, compute, and long-term commitment behind the initiative.

Apply now if this sounds like a great fit!

*Applications remain confidential*

*No contractors*