Staff Engineer, Physical AI & Video Modeling

Build AI; San Francisco (CA); On-site; USD 230,000 - 260,000; Full time

Benefits offered by this job

Competitive pay

Medical, dental, and vision packages

Housing subsidy

Relocation support

Wellness benefits

Daily lunch and dinner

Unlimited compute budget

Unlimited Codex and Claude credits

Travel

Job summary

Build AI in San Francisco (Financial District) is seeking Members of Technical Staff to pursue bold research bets on physical-labor, video, and world-model data, with access to a scale dataset and a small team that lets you train on it.

You will work across research and engineering to deliver practical results, designing experiments, training models with PyTorch, and shaping scalable data pipelines. This role blends exploration with engineering execution in a hands-on, in-person team.

Qualifications

• You’ve trained real models and can defend a research bet.
• Proficiency in Python and a deep learning framework (PyTorch or equivalent).
• You want to work on in-the-wild data, not only academic splits.
• You can move between research and engineering to complete experiments.

Responsibilities

• Run research bets on Build data: train, evaluate, ablate, and ship what works.
• Design and train models on in-the-wild video and physical-labor data.
• Build training and data paths (PyTorch, distributed training, dataset curation).
• Collaborate with Head of Dataset & Quality to drive results.
• Cover research and engineering in a small team (under 30).
• Design evaluation frameworks that reflect generalization, not just training curves.

Skills

Python

Deep learning

Research mindset

Distributed training

Independent work

Tools

PyTorch

TensorFlow

CUDA

Job description

Build AI in San Francisco (Financial District) is seeking Members of Technical Staff to pursue bold research bets on physical-labor, video, and world-model data, with access to a scale dataset and a small team that lets you train on it.

You will work across research and engineering to deliver practical results, designing experiments, training models with PyTorch, and shaping scalable data pipelines. This role blends exploration with engineering execution in a hands-on, in-person team.