Machine Learning Engineer - Perception

Job Summary

As a Machine Learning Engineer – Perception at CurvePoint, you will design, train, and optimize perception models that power our Wi-AI Core technology — an advanced sensing platform that uses wireless signals to detect and identify people, objects, and threats in real-world environments. Your work will focus on computer vision–inspired deep learning for time-series signal data, with opportunities to bridge traditional CV techniques into novel RF-based sensing. You will work closely with our research scientists to create robust, real-time models that perform reliably across varied, dynamic, and noisy environments.

Duties and Responsibilities

• Model Development – Design and train machine learning models for detection, classification, and tracking from complex, multi-modal sensor inputs (e.g., wireless CSI, video, other signals).
• Data Pipeline Engineering – Build and maintain scalable data preprocessing, augmentation, and labeling pipelines tailored to time-series and perception datasets.
• Model Optimization – Experiment with architectures (CNNs, transformers, 3D vision models) to improve robustness and generalization across environments.
• Testing & Validation – Develop evaluation protocols, benchmark performance, and iterate quickly to address failure modes.
• Collaboration with Hardware – Work with hardware engineers to understand sensor capabilities and ensure alignment between data collection and model needs.
• Deployment & Integration – Package and deploy models to production, including embedded or edge inference environments (e.g., NVIDIA Jetson).
• Research Translation – Stay up-to-date on the latest CV and perception research and rapidly integrate relevant advances into production workflows.

Must Haves

• Bachelor’s or Master’s degree in Computer Science, Electrical Engineering, Robotics, or related field, with a focus on computer vision, machine learning, or signal processing.
• 3 years of experience developing deep learning models for perception (academic or industry).
• Strong skills in Python and familiarity with PyTorch (or TensorFlow), OpenCV, and data pipeline frameworks.
• Experience with model training, hyperparameter tuning, and deployment in real-time or embedded environments.
• Solid understanding of model evaluation metrics, overfitting prevention, and generalization strategies.

Nice to Have

• Experience with RF or wireless sensing, especially CSI-based data or other non-visual perception modalities.
• Background in sensor fusion (e.g., combining RF, video, LiDAR, or radar inputs).
• Experience with domain adaptation, transfer learning, or synthetic data generation for perception tasks.

Location: Pittsburgh, PA

Salary: $140,000 - $170,000/year