Senior Production ML Engineer – Vision & Multimodal Systems

Dex; Greater London; On-site; GBP 90,000 - 150,000; Full time

Job summary

Dex, in partnership with trusted collaborators, seeks an experienced engineer to own architecture, deployment, and reliability of vision and multimodal models in production. This role focuses on systems-first software engineering to solve GPU utilisation, latency, and throughput bottlenecks for real-time inference at scale.

You will join the in-house ML team in London, on-site four days a week, shaping end-to-end model serving and integration within a scalable platform.

Qualifications

• 5+ years of software engineering experience with deep expertise building and running production ML systems or model-backed distributed infrastructure at scale.
• Hands-on mastery of deep learning model serving and inference optimisation, including GPU utilisation, dynamic batching, and latency-throughput-cost trade-offs.
• Proven track record designing and operating scalable distributed systems on Kubernetes and cloud infrastructure (GCP/AWS) using Python and PyTorch.
• Demonstrated ownership taking experimental research prototypes and re-architecting them into dependable, maintainable platform features.

Responsibilities

• Architect and run low-latency serving infrastructure for heavy vision and multimodal models, delivering real-time interactive predictions directly into high-throughput annotation workflows.
• Build scalable offline inference pipelines processing millions of hours of egocentric video for robotics customers, balancing GPU cluster costs, throughput, and fault tolerance.
• Productionise agentic quality-control and active learning workflows, transforming fragile research prototypes into resilient, self-healing platform features.
• Design dynamic model integration layers that gracefully handle rapid, customer-defined ontology shifts without requiring bespoke one-off code.
• Profile and optimise end-to-end model serving performance across PyTorch, CUDA, and Kubernetes environments on AWS and GCP.

Job description

Dex, in partnership with trusted collaborators, seeks an experienced engineer to own architecture, deployment, and reliability of vision and multimodal models in production. This role focuses on systems-first software engineering to solve GPU utilisation, latency, and throughput bottlenecks for real-time inference at scale.

You will join the in-house ML team in London, on-site four days a week, shaping end-to-end model serving and integration within a scalable platform.