About the role
Type: Full-time Entry Level Role
Start date: ASAP
Location: SF, USA, remote — On-site/hybrid possible
The successful candidate must be eligible to work in the US or the UK. We are not able to sponsor visas at this time.
The Role
Netholabs is building AI grounded in biological intelligence. We record petascale, high-resolution neurobehavioural data from living systems and use it to train neural foundation models — a new substrate for the next generation of AI, robotics, and personalized intelligence.
We're looking for an AI/ML engineer to support our research and engineering team across model training, data pipelines, and applied ML work. You'll get hands-on exposure to how a neural foundation model is actually built — from raw neurobehavioural data through to training runs and downstream applications in robotics and embodied AI. This is a broad, hands-on role: you'll work closely with our research engineers, take on real pieces of active projects, and grow into the areas that fit you best.
Responsibilities
Model Development
Support training, fine-tuning, and evaluation of neural foundation models
Run experiments, track results, and help iterate on model architectures
Work with biomechanical data, machine vision data, and robotics control problems
Help benchmark model performance and write up findings
Data & Infrastructure
Build and maintain data pipelines for petascale neurobehavioural datasets
Clean, preprocess, and structure multi-modal data (video, sensor, physiological) for training
Help keep experiment tracking, datasets, and compute usage organized
Applied ML & Robotics
Support applications of trained models to robotics and embodied-agent tasks
Prototype small tools, scripts, and demos to test model capabilities
Contribute to internal documentation as work progresses
Requirements
Core (essential)
Strong Python skills and comfort working in a Linux/command-line environment
Solid foundation in ML fundamentals (e.g., through coursework, projects, or research)
Experience with at least one deep learning framework (PyTorch preferred)
Experience training ML models on time-series datasets
Curious, self-directed, and comfortable working with ambiguity in a fast-moving research environment
Good communication; able to document work clearly as you go
Valued (or willing to learn)
Experience with machine vision or robotics problems
Exposure to large-scale model training or distributed compute
Experience with data pipelines, structured storage, or large dataset handling
Familiarity with robotics, sensorimotor learning, or embodied AI
Background in neuroscience, behavioural science, or related fields
No prior neuroscience or robotics experience required; we will cross-train the right person.