LLM Inference Platform Architect

JoinQuTwo - Europe's private Quantum AI Lab, on our quest to bridge the gap between the classical and the quantum world. We partner with the most advanced and forward-thinking customers, solving fundamental real-world problems, side by side.

We are a European AI lab for the quantum era, co-founded by pioneers behind IQM and Silo AI, with a team spanning MIT, UC Berkeley, Cambridge, and Carnegie Mellon, and backed by the founders of Skype, Hugging Face, Quantinuum, Supercell and many more. We work at the intersection of quantum computing, AI, optimization, and simulation - where the answers do not yet exist and the work genuinely matters.

We are seekingLLM Inference Platform Architect. If you are a humble learner, not afraid of the unknown, we encourage you to apply.

Role Description

The Infrastructure Architect designs and leads a secure, internal LLM serving platform on fully controlled GPU hardware, ensuring complete data privacy and independence from external model providers.

Managing the full serving stack-including GPU clusters, inference engines, access control, and capacity planning-this role delivers a reliable AI service for the entire organization. The architect sets platform standards, ensures strict regulatory compliance (such as DORA), and partners closely with security, operations, and LLM scientists to deploy optimized models.

Qualifications

• Experience designing and running large-scale infrastructure on Kubernetes, including GPU workloads

• Hands-on experience with LLM inference engines (vLLM, SGLang, TensorRT-LLM or similar)

• Infrastructure as code (Terraform or similar), CI/CD and observability tooling

• Understanding of security, identity and network design in enterprise environments

• Experience writing architecture guidelines and working with enterprise architecture teams

Experience in one or more of the following areas is a plus but not a requirement:

• Hands-on experience configuring and managing enterprise-grade GPUs (NVIDIA or AMD Instinct) and their respective acceleration stacks (CUDA or ROCm).

• Background in banking or other highly regulated industries, with familiarity with DORA or equivalent compliance frameworks.