Software Inference Deployment Engineer at Lumai

Department: Product

Setup: Hybrid
Location: Oxford, England
Level: mid
Posted: Jun 17, 2026

Description

Lumai is hiring a Software Inference Deployment Engineer to integrate and support its Iris optical AI compute platform in third-party data centre environments. The role begins with software-stack integration, model onboarding, and familiarity with the disaggregated prefill/decode runtime, then moves to customer integration, troubleshooting, training, and feedback. Candidates need software engineering experience in AI infrastructure or accelerator integration, strong Python and PyTorch skills, model-deployment experience, inference-serving familiarity, and Linux, Docker, and cluster experience. The position offers competitive salary, share options, pension, private health insurance, development allowance, subsidised lunches, and 25 paid holiday days.

Required Skills

Python
PyTorch
Inference serving
Linux
Docker
Cluster environments
Model onboarding

Additional Information

Required Previous Experiences: Hands-on software engineering experience in AI infrastructure, inference serving, accelerator integration, or comparable deep-tech hardware-software environments
Practical experience with model deployment workflows, including loading, format conversion, quantisation, or framework integration
Experience working with inference serving stacks such as vLLM or TensorRT-LLM
Preferred Qualifications: Experience integrating accelerator hardware into customer inference workflows
Familiarity with the NVIDIA inference stack, including CUDA, TensorRT, and Triton
Exposure to disaggregated inference architectures, prefill/decode separation, or KV cache management