Machine Learning Sde, Scanless Technologies

What you'd actually do

1. Design, build, and maintain end-to-end solutions that automate data collection, processing, annotation, model training, validation, and deployment for computer vision models operating across thousands of edge devices in production.
2. Collaborate with applied scientists and ML engineers to operationalize research models into production-ready pipelines, bridging the gap between offline experimentation and real-world deployment on constrained edge hardware.
3. Partner with hardware and optics teams to validate sensor configurations, calibration accuracy, and image quality requirements that directly impact model performance, closing the feedback loop between field hardware and ML training.
4. Participate in on-call rotations, triaging ML pipeline failures and model performance degradations in production, performing root cause analysis, and driving resolution to maintain fleet-wide model health.
5. Implement automated test frameworks including unit, integration, stress, hardware-in-the-loop, and long-running reliability test suites that validate end-to-end system behavior across cloud and edge boundaries before every production deployment

Skills

What the JD emphasized

• real-world Robotics
• computer vision
• Machine Learning
• constrained edge hardware
• model performance
• model health
• production deployment

Other signals

• Deploying ML models on edge devices
• Building ML pipelines for computer vision models
• Operationalizing research models into production