Software Engineer - ML/Computer Vision (Battery Sorting)

Redwood is localizing a global battery supply chain that seamlessly integrates recovery, reuse, and recycling — keeping critical minerals in circulation and driving the energy transition. Founded in 2017, we’re delivering low-cost and large-scale energy storage and producing battery materials in the U.S. for the first time, all from batteries we already have.

Software Engineer, ML/Computer Vision (Battery Sorting)

The Battery Sorting team at Redwood Materials is building a world-class, ML-enabled sorting platform that uses computer vision and machine learning to classify and route thousands of end-of-life batteries per hour across diverse chemistries and form factors. This role sits at the intersection of software engineering and machine learning, with direct ownership of the production systems powering automated battery sorting on the factory floor. The ideal candidate is equally comfortable debugging a production incident as iterating on a model, and will have the opportunity to generate patents in automated battery classification. This is a high-impact, highly visible role with immediate real-world application in advancing the energy transition.

Hours

Full-time | Schedule may vary depending on site operational needs; flexibility required

• Develop, test, and maintain production software systems powering automated battery sorting, spanning ML inference, image acquisition, sensor integration, and hardware-adjacent control interfaces

• Train and deploy computer vision models for battery chemistry classification, including dataset annotation, preprocessing, and evaluation within established data pipelines

• Build and maintain services and APIs that connect ML outputs to downstream systems including MES, HMI, and PLC/controls interfaces

• Own observability across the production software stack through structured logging, metrics dashboards, alerting, and on-call triage for inference pipelines and supporting services

• Monitor model performance in production to catch regressions or distribution shifts and drive iterative improvements through data analysis and retraining

<li