ML/CV Engineer

ML/CV Engineer

Organisation Overview

Our Client is developing advanced AI technology for the healthcare sector, helping modernise surgical and procedural environments through computer vision, intelligent automation and real-time operational insight.

The organisation combines research-led innovation with practical deployment, creating systems designed to operate safely and reliably in demanding clinical settings. Employees benefit from a collaborative engineering culture, opportunities for continuous development, comprehensive healthcare options, paid leave and access to a retirement plan.

Role Summary

Our Client is seeking an ML/CV Engineer to take computer vision and deep learning models from successful prototypes into production on embedded hardware used in hospital environments.

This is a high-impact position with ownership extending beyond initial deployment. You will optimise models for real-world hardware constraints, establish robust retraining and release processes, monitor performance in the field and manage carefully controlled updates across deployed devices.

Working alongside research, embedded software and quality engineering teams, you will help turn advanced machine learning capability into dependable, scalable products used within critical healthcare workflows.

Responsibilities

• Deploy computer vision and deep learning models to embedded accelerators, applying techniques such as quantisation, pruning, operator fusion and graph optimisation.
• Analyse inference performance on target hardware and improve kernels, memory use, batching and processing schedules to meet defined latency and throughput requirements.
• Develop and maintain reproducible pipelines for model training, evaluation, packaging and release.
• Create automated regression testing processes using appropriately held-out clinical datasets.
• Add monitoring for inference latency, resource consumption and prediction behaviour across deployed systems.
• Investigate model drift, unexpected behaviour and performance deterioration on devices in the field.
• Plan and execute phased model releases, including shadow testing, approval criteria and rollback processes.
• Manage retraining programmes from data selection and labelling requirements through to validation and release evidence.
• Collaborate with research engineers to make experimental models suitable for production use.
• Work with embedded software teams to support efficient hardware deployment.
• Partner with quality engineering to prepare documentation and evidence for regulated product releases.
• Maintain ownership of model performance and lifecycle management following deployment.

Essential Skills & Experience

• At least four years’ professional experience developing and deploying computer vision or deep learning models.
• Experience taking at least one machine learning model into production on embedded or edge hardware.
• Strong production programming experience using both Python and C++.
• Practical knowledge of inference runtimes and optimisation tools, such as TensorRT, ONNX Runtime, OpenVINO, TFLite or comparable embedded accelerator SDKs.
• Experience applying model compression methods, including post-training quantisation, quantisation-aware training, pruning and knowledge distillation.
• Understanding of how optimisation and compression techniques affect model accuracy and operational performance.
• Experience working with embedded compute platforms, such as NVIDIA Jetson, Arm-based systems, GPUs, DSPs or NPUs.
• Ability to profile and troubleshoot machine learning workloads on target hardware.
• Demonstrable experience managing models after release, including version control, monitoring, retraining and controlled production rollout.
• Candidates must be based in the United States and already authorised to work there.
• Our Client is unable to provide employment visa sponsorship now or in the future.

Desirable Skills & Experience

The following would strengthen an application but are not essential:

• Experience delivering software within a regulated medical device environment.
• Familiarity with design controls, verification and validation activities.
• CUDA or other GPU kernel development for custom operators.
• Experience with video capture, real-time streaming or hardware-accelerated video processing on embedded platforms.
• Knowledge of MLOps tools for experiment tracking, dataset versioning and model registries.
• Experience with containerised deployment and CI/CD for embedded or fleet-managed devices.
• A postgraduate qualification in computer science, electrical engineering or a related discipline.

Apply Now

If you have proven experience deploying and maintaining computer vision models on embedded hardware, please submit your CV for consideration. This opportunity is particularly suited to engineers who enjoy bridging research, software engineering and reliable real-world deployment.

Disclaimer: Any remuneration stated in this advertisement is indicative and will be dependent on the successful candidate’s experience, skills and qualifications. Final remuneration will be agreed with the successful candidate and will be subject to the terms of the contract.