← All open roles
Machine Learning & Signal Processing Engineer
ScienceFull-time · OnsiteIIT Madras, Chennai
You'll build the science behind Orbit's scores — turning noisy, real-world biosignals (EEG, PPG/HRV) into reliable measures of focus, load, and recovery. It's hands-on work across filtering, feature extraction, and modeling, with a direct line from your code to what people see in the app.
This is an on-site role at our Chennai facility, working closely with our neuroscience and product teams.
What you'll do
- Design biosignal pipelines — filtering, artifact removal (ICA/PCA, wavelets), and feature extraction across time, frequency, and time-frequency domains.
- Train and validate ML/DL models for classification and regression on EEG, ECG, and PPG/HRV data.
- Move models from notebook to real-time inference that runs on the wearable.
- Partner with neuroscience and product to turn research questions into measurable, trustworthy scores.
What you'll bring
- Hands-on experience with biosignal processing, ideally EEG and ECG.
- A strong grounding in signal processing, plus linear algebra, probability, multivariate statistics, and optimization.
- Practical ML/DL experience applied to biosignals — both classification and regression.
- Fluency in Python and the ecosystem: scikit-learn, TensorFlow/PyTorch, Pandas, NumPy, and MNE.
Bonus points
- Real-time biosignal pipelines for wearables or neurotech.
- PPG/HRV experience.
- Comfort with cloud platforms (Colab, AWS, Azure) for training and deployment.
Why Neurostellar
You'll build at the intersection of neuroscience, design, and engineering — on products people use to understand and train their own minds. Early-stage means real ownership: your work ships, and you watch it land in people's hands.
Neurostellar is an equal-opportunity employer. We value diverse perspectives and welcome applicants from every background.