remote
Applied Scientist - Workforce Safety Data Tech - Amazon.com
Software Engineer
Lead the design, training, and deployment of computer‑vision and machine‑learning models that enhance workforce safety across Amazon’s global operations, leveraging video, image, and sensor data to detect anomalies, recognize activities, and predict risks.
About the role
Key Responsibilities
- Design, develop, and train state‑of‑the‑art computer‑vision models for activity recognition, anomaly detection, object detection, and risk prediction using video, image, and sensor data.
- Collaborate with software engineers to transition research prototypes into scalable, production‑ready services deployed across hundreds of Amazon facilities worldwide.
- Conduct rigorous experimentation, hyper‑parameter tuning, and model evaluation to ensure high accuracy and robustness in diverse operational environments.
- Integrate multimodal data streams and develop pipelines for real‑time inference and continuous model monitoring.
- Communicate findings, model performance, and impact metrics to cross‑functional stakeholders, translating technical insights into actionable safety improvements.
Requirements
- Ph.D. or Master’s in Computer Science, Electrical Engineering, or related field with strong emphasis on machine learning and computer vision.
- Proven experience building and deploying deep learning models at scale, preferably in a production setting.
- Proficiency in Python and deep learning frameworks such as TensorFlow or PyTorch.
- Solid understanding of video analytics, anomaly detection, and risk prediction techniques.
- Excellent problem‑solving skills, ability to work independently, and strong communication abilities.
Skills
computer visionmachine learningpythondeep learningtensorflow