Lead AI Inference Engineer

AI Connect | Data & AI Delivery Partner; City of Edinburgh; Remote; GBP 110,000 - 160,000; Full time

Job summary

AI Connect | Data & AI Delivery Partner is seeking a Lead AI Inference Engineer to lead production-grade inference pipelines in a fully remote role based in the UK. You will own model hosting, serving, training and fine-tuning workflows across GPU infrastructure and Kubernetes, driving latency, throughput and cost optimisations at scale.

You will collaborate with ML and AI engineering teams to onboard models, automate deployment, monitor systems and rollback strategies, ensuring reliability and

Qualifications

• Strong Python and production software/ML systems engineering experience.
• Hands-on experience with GPU/ML infrastructure, model serving or inference optimisation.
• Knowledge of Kubernetes, cloud infrastructure and automation (AWS preferred).
• Exposure to technologies such as vLLM, Triton, KServe or similar advantageous.
• Experience building reliable, scalable systems with strong engineering standards.
• Inference performance: batching, caching, quantisation, benchmarking and latency.

Responsibilities

• Building and operating model hosting, serving and deployment capabilities.
• Developing scalable GPU-backed infrastructure and inference services.
• Improving model latency, throughput and cost through performance optimisation.
• Supporting model onboarding, deployment automation, monitoring and rollback.
• Working closely with ML and AI Engineering teams to bring models into production.

Skills

Python

GPU/ML infrastructure

Kubernetes

Cloud infrastructure

Automation

Tools

AWS

vLLM

Triton

KServe

Job description

Lead AI Inference Engineer: Fully Remote (UK Based)

Six Figure Salary + Package

The client is a major UK financial institution building a new AI Engineering capability to support customer and colleague experiences at significant scale. The opportunity involves real production engineering, complex governance and collaboration across global engineering teams.

Lead engineering of shared model hosting, serving, training and fine-tuning capabilities for approved internally managed and open-source models. Work across GPU infrastructure, Kubernetes, serving frameworks, model packaging, inference optimisation, capacity planning and operational reliability.

What you'll be doing

• Building and operating model hosting, serving and deployment capabilities.
• Developing scalable GPU-backed infrastructure and inference services.
• Improving model latency, throughput and cost through performance optimisation.
• Supporting model onboarding, deployment automation, monitoring and rollback.
• Working closely with ML and AI Engineering teams to bring models into production.

What we're looking for

• Strong Python and production software, platform or ML systems engineering experience.
• Hands-on experience with GPU/ML infrastructure, model serving or inference optimisation.
• Knowledge of Kubernetes, cloud infrastructure and automation, ideally AWS.
• Exposure to technologies such as vLLM, Triton, KServe or similar would be advantageous.
• Experience building reliable, scalable systems with strong engineering standards.
• Inference performance: batching, caching, quantisation, benchmarking and latency