Head of AI, Surgical Vision & Robotics Platform

Dex; Greater London; Sur place; GBP 120 000 - 170 000; Plein temps

Résumé du poste

Dex is partnering with a company building a surgical intelligence platform that combines computer vision, AI, and AR to enable real-time navigation for surgeons and robots. You will lead the AI pillar end-to-end, building a team from the ground up to ship production models and drive federated learning and MLOps at scale.

The role focuses on delivering an anatomy recognition model for neurosurgery by Q2 2028, advancing the foundation model toward vision-language-action for device automation, and

Qualifications

• Shipped production AI/ML models.
• CV/segmentation or vision-language experience is ideal.
• Experience with medical imaging (DICOM, MONAI) or rapid ramp-up.
• Dealt with real-time constraints: sub-20ms latency and edge deployment.
• Led 2-3 engineers and built a full AI pillar.
• Based in or near London; able to work in person 3+ days a week.

Responsabilités

• Deliver end-to-end AI pillar for the surgical platform from CV/segmentation to vision-language-action.
• Architect data pipelines for kinematics and tissue-interaction data across endpoints.
• Own the federated learning stack; deploy NVIDIA FLARE for privacy-preserving improvements.
• Collaborate with the simulation team on Isaac Sim, synthetic data, and sim-to-real transfer.
• Ship production anatomy recognition model for neurosurgery by Q2 2028 with ≥85% accuracy.

Connaissances

Production AI/ML

Computer vision

Medical imaging

Real-time inference

Federated learning

MLOps

Team leadership

Edge deployment

Formation

Bachelor's degree in a related field

Outils

NVIDIA FLARE

DGX/RTX infrastructure

MONAI

Isaac Sim

NVIDIA AGX

Description du poste

Dex is partnering with a company building a surgical intelligence platform that combines computer vision, AI, and AR to enable real-time navigation for surgeons and robots. You will lead the AI pillar end-to-end, building a team from the ground up to ship production models and drive federated learning and MLOps at scale.

The role focuses on delivering an anatomy recognition model for neurosurgery by Q2 2028, advancing the foundation model toward vision-language-action for device automation, and