Machine Learning Engineer

Recycleye•Stratford, CT

18d•Hybrid

About The Position

Recycleye is hiring a Machine Learning Engineer to develop features for its waste-sorting and plant-monitoring products. The role is part of a small, high-impact ML R&D team, working on production systems installed on real sorting lines. The engineer will evaluate and deploy model architectures, adapt machine-learning research to practical use cases, implement object-tracking algorithms, work with camera optics and calibration, and improve DeepStream pipelines. Solutions should be selected for their effectiveness in real-world conditions, including situations where a hardware or camera change may be more appropriate than a new model. The role involves building product features and better ways to measure performance, collaborating with the data team and colleagues developing a plant-monitoring product, and supporting deployed systems. The engineer will spend time at customer sites and the demo centre, working directly with sorting equipment and waste. As the team grows, the engineer will help shape processes such as code review and testing and take on increased responsibility for planning and technical decisions. The position uses primarily Python, with some Go, and deploys models and algorithms to NVIDIA Jetson devices running DeepStream pipelines, supported by Docker, GitHub Actions and Azure. Success means regularly delivering work to production with measurable impact, bringing and testing new ideas, growing into ownership of technical areas, and becoming a trusted reviewer and planning partner. The engineer will work with teams across the company and contribute to improving how system performance is measured and how products improve with each deployment. The role is hybrid, based in Stratford E15 2HU, with occasional travel to Europe or the US.

Requirements

Hands-on computer vision experience and a background in machine learning, physics, maths, engineering or computer science.
Around 3–5 years’ experience writing production-quality code, primarily in Python, and promoting good engineering practice.
A track record of shipping machine-learning systems into production in a commercial setting; computer vision on edge devices is especially relevant.
Solid software engineering fundamentals, including tested code, version control and code review.
Clear communication with technical and non-technical colleagues and a collaborative approach.
Comfort working independently and as part of a small team.
Proactive, driven and practical, comfortable with change, and keen to learn.
Creative and practical problem-solving skills.
Willingness to spend time on site and at the demo centre, testing in challenging environments and working hands-on with waste.

Nice To Haves

Experience with computer-vision detector-classifier models such as YOLO or RF-DETR, and segmentation models.
Experience with multi-object tracking, camera calibration or homography transforms.
Experience deploying to NVIDIA Jetson and using DeepStream or GStreamer.
Experience with Docker, CI/CD such as GitHub Actions, or cloud platforms such as Azure, AWS or GCP.
Experience with model evaluation and monitoring in production.
Experience with Go.
Experience building tools used by other teams.
Experience with physical systems or robotics, cameras and lenses.
Commercial awareness, including understanding how technical choices affect customers and the business.

Responsibilities

Develop features for Recycleye’s waste-sorting and plant-monitoring products.
Evaluate and deploy model architectures and adapt machine-learning research to practical use cases.
Implement algorithms, including object-tracking methods, and work on camera optics and calibration.
Improve the efficiency of DeepStream pipelines and deploy models and algorithms to production systems.
Develop ways to measure product performance, including sorting performance and the effects of real-site conditions.
Collaborate with the data team and colleagues developing the plant-monitoring product.
Help support deployed systems and spend time on site and at the demo centre testing in challenging environments.
Contribute to team practices, including code review and testing, and take on more responsibility for planning and technical decisions.
Share learning about computer vision and the waste-sorting industry across the company.
Help teams across the company get the best from Recycleye’s systems.