Research Engineer

The role

You will design, implement, and run pre-training and post-training pipelines for action-conditioned world models and vision-language-action models. You will build distributed training infrastructure and work with multimodal data pipelines.

What you will do

• Design, implement, and run pre-training and post-training pipelines for action-conditioned world models and vision-language-action (VLA) models
• Develop and refine training methodologies, including fine-tuning, reinforcement learning, and large-scale multimodal learning
• Design and generate training and evaluation datasets from simulation
• Build distributed training infrastructure using PyTorch, FSDP, and DeepSpeed
• Work with multimodal data pipelines involving video, sensory inputs, and action sequences
• Evaluate model performance using both benchmark datasets and real-world deployment metrics
• Collaborate with industrial partners to adapt generative models for real-world physical AI applications

Required qualifications

• Experience with pre-training or post-training on large generative models (video, multimodal, or action-conditioned)
• Hands-on proficiency with PyTorch and distributed training frameworks (FSDP, DeepSpeed)
• Strong fundamentals in machine learning, optimization, and large-scale data processing
• Familiarity with VLMs, VLAs, or world models
• BS/MS/PhD in Computer Science, Machine Learning, Robotics, or a related field

Nice to have

• Contributions research publications
• Background in robotics, embodied AI, or sim-to-real transfer
• Experience with video understanding or temporal reasoning

Benefits

• Competitive compensation and equity
• 401k (no match)
• Healthcare (Silver PPO Medical, Vision, Dental)
• Lunch and snacks at the office

About the company

Hedra is a research and infrastructure lab advancing post-training and inference for multimodal and physical AI. The team works fully in person in San Francisco.