Machine Learning Engineer

About the Role 
We’re looking for a Machine Learning Engineer passionate about time-series and sensor data to help transform raw signals into deployable models for our fall detection MVP and beyond. Your work will power gait-based recognition, anomaly detection, and health-focused automation features. 

Key Responsibilities 

• Preprocess and clean multi-sensor data (mmWave radar, ZED depth, IMUs). 

Experiment with time-series deep learning models (LSTMs, Transformers, TCNs) for fall detection. 

Build unique human identification models using gait and skeletal signatures. 

Train/test/evaluate using both public datasets (MHAD, HCA) and Arqaios-collected data. 

• Optimize and deploy models for edge environments (TensorRT, TF Lite, ONNX). 
•  

Minimum Experience: 2–3 years in applied ML, preferably with time-series or sensor data. 

Entry bar: 2 years working with deep learning models (LSTMs, Transformers, or HAR-related). 

Stronger candidates: 3–5 years in applied ML engineering with experience in deploying ML to production (cloud or edge). 

Required Skills & Qualifications 

Proficiency in Python, PyTorch/TensorFlow, and sklearn. 

Experience with time-series modeling, HAR, or anomaly detection. 

Knowledge of data preprocessing, feature engineering, and augmentation. 

Familiarity with edge inference frameworks. 

Preferred / Plus Points 

Background in sensor data (radar, IMU, LiDAR, or similar). 

Research or publications in HAR, biometric identification, or fall detection. 

Experience deploying ML in resource-constrained or IoT environments. 

Why Join Us? 
Your models will be the intelligence layer of a system that makes interior spaces safer and smarter. This is not just research — your work will directly impact real-world lives.