Senior AI Application Engineer - LLM & VLM Enablement - Hybrid/Remote - Europe (incl. UK), United Kingdom

Originally posted 26 October 2026 by the employer.

About the role

As a Senior AI Application Engineer in the Model Enablement team (Application Software), you will own LLM and VLM support on Metis and Europa AIPUs, from studying new models to shipping them in the Voyager SDK.

What you'll do

- Study new LLM and VLM families and architectures (such as mixture-of-experts, small VLMs, and new attention and decoding schemes), prototype and profile them on Metis and Europa, and decide how they run on hardware. - Turn internal ML research results into supported Voyager SDK features with stable APIs, tests, documentation, and release gates. - Own PyTorch reference implementations with built-in quantization, weight mapping, and contracts detailing inputs, outputs, weights, and persistent state like paged KV cache for the compiler team. - Integrate LLM and VLM serving into the Voyager SDK pipeline builder, including prefill and decode, KV-cache management, streaming text output, and multimodal inputs. - Measure accuracy against floating-point references alongside regression and quantization owners to resolve or explain performance gaps. - Extend the model enablement platform for LLM and VLM bring-up with compiler, runtime, and hardware-in-the-loop environments.

What you'll need

- 5+ years of software engineering experience, with at least 2 years focused on ML inference, ML systems, or ML tooling. - Solid understanding of transformer inference, including attention variants, KV caching, prefill versus decode, sampling, tokenization, and vision encoder integration in VLMs. - Working knowledge of LLM quantization schemes (weight-only, weight-activation, calibration, outlier handling) and accuracy analysis. - Strong Python and PyTorch proficiency with experience shipping production-quality software. - Experience using AI coding agents on production codebases to build, test, and verify implementations. - Demonstrated ability to convert research code and hardware constraints into practical software designs.

Nice to have

- Experience with inference engines or serving stacks such as llama.cpp, vLLM, ONNX Runtime GenAI, or ExecuTorch. - Experience with kernel development, tensor compilers (such as Triton, TVM, or MLIR), or profiling models on accelerator hardware. - Experience with speech (ASR, TTS) or agentic applications on embedded devices. - Experience building evaluation and testing harnesses for ML systems, AI agents, or tool-calling agents such as MCP servers.

Skills: LLM inference, Voyager SDK, quantization, KV cache, PyTorch