P
Preload Verified
Biosignal ML Engineer / Data Scientist
San Francisco, CAOnsiteFull Time$150,000–$225,000 /yrPosted 1 day agoHidden Gem · YC Startup
Required skills
- biosignals
- time-series data
- EMG
- tactile
- kinematic
- Python
- ML fundamentals
- PyTorch
- signal processing
- filtering
Nice to have
- multimodal learning
- self-supervised learning
- neuroengineering
- BME
- robotics
Role summary
Preload is seeking a Biosignal ML Engineer / Data Scientist to join their early-stage, in-person team in San Francisco. This role involves transforming raw biosignal data (EMG, tactile, kinematic) into features and models for physical AI. Responsibilities include building extraction pipelines, time-aligning and fusing multimodal data streams, developing models for human manipulation and motor intent, defining data quality metrics, and collaborating with hardware and robotics teams. The ideal candidate has 2+ years of experience with time-series or biosignal data, strong Python and ML fundamentals (PyTorch), and solid signal processing skills.
**About Preload**
Preload builds the data layer for physical AI. Robots learn mostly from video, but human interaction is much richer: touch, force, and muscle activity. We build a wearable multimodal stack (egocentric video, tactile gloves, EMG) to capture how people actually interact with the physical world, and turn it into datasets robotics teams can train on. We're a YC (F26) team of PhD founders from Johns Hopkins and Caltech with backgrounds at Neuralink, Paradromics, and other neurotech companies.
**The role**
You'll turn raw, noisy biosignals into features and models that make our data useful for physical AI.
**What you'll do**
- Build feature extraction pipelines for EMG, tactile, and kinematic signals
- Time-align and fuse multimodal streams (video, touch, muscle activity)
- Develop models and representations of human manipulation and motor intent
- Define data quality metrics and help shape what we collect next
- Work closely with sensor and robotics teammates to close the loop from hardware to model
**What we're looking for**
- 2+ years working with time-series or biosignal data (EMG, EEG, IMU, neural recordings, or similar)
- Strong Python and ML fundamentals (PyTorch or similar)
- Solid signal processing skills: filtering, artifact removal, synchronization
- Experience with multimodal or self-supervised learning is a plus
- Background in neuroengineering, BME, robotics, or related fields is a plus
**Why join**
Early, in-person team in San Francisco. You'll own core technical decisions from day one.
Preload builds the data layer for physical AI. Robots learn mostly from video, but human interaction is much richer: touch, force, and muscle activity. We build a wearable multimodal stack (egocentric video, tactile gloves, EMG) to capture how people actually interact with the physical world, and turn it into datasets robotics teams can train on. We're a YC (F26) team of PhD founders from Johns Hopkins and Caltech with backgrounds at Neuralink, Paradromics, and other neurotech companies.
**The role**
You'll turn raw, noisy biosignals into features and models that make our data useful for physical AI.
**What you'll do**
- Build feature extraction pipelines for EMG, tactile, and kinematic signals
- Time-align and fuse multimodal streams (video, touch, muscle activity)
- Develop models and representations of human manipulation and motor intent
- Define data quality metrics and help shape what we collect next
- Work closely with sensor and robotics teammates to close the loop from hardware to model
**What we're looking for**
- 2+ years working with time-series or biosignal data (EMG, EEG, IMU, neural recordings, or similar)
- Strong Python and ML fundamentals (PyTorch or similar)
- Solid signal processing skills: filtering, artifact removal, synchronization
- Experience with multimodal or self-supervised learning is a plus
- Background in neuroengineering, BME, robotics, or related fields is a plus
**Why join**
Early, in-person team in San Francisco. You'll own core technical decisions from day one.
