AI Inference Engineer (all genders)

Frankfurt am Main, Karlsruhe, Leipzig, Mannheim, Stuttgart

• You design, develop and operate productive LLM inference platforms for customers with high demands for data sovereignty, security and operational control – on-premises, in private cloud environments or sovereign European cloud setups.
• Together with cloud, platform, security and data engineering teams, as well as our customers, you will bring AI use cases into productive operation.
• You will integrate modern inference engines and open-weights models into Kubernetes, container, and platform environments.
• Furthermore, you plan and optimize GPU and memory resources as well as inference workloads: from model sizes, quantization and batching to KV cache strategies, latency, throughput and costs.
• You are responsible for the runtime of production AI systems, including model serving, APIs, authentication, secrets, observability, and logging.
• From customer projects, you will develop reusable reference architectures, deployment templates and operational playbooks, thereby strengthening our Applied AI capability.

• Personal background: Experience in platform engineering, cloud infrastructure, MLOps, LLMOps, DevOps, backend engineering, or machine learning engineering. Crucially, you have experience building and operating production systems and a strong drive for rapid personal development.
• Inference Engineering: You understand the technical and economic aspects of modern LLM inference, from model serving and GPU utilization to quantization, batching and KV cache management, as well as latency, throughput and costs.
• Cloud & Platforms: Docker, Kubernetes, Helm, Terraform, CI/CD, Linux and Observability are part of your daily work routine.
• AI understanding: You can classify transformer-based models such as LLMs and embeddings and make informed technical decisions for productive AI systems.
• Security & Governance: You consider topics such as identities, permissions, secrets, logging, auditing and compliance from the very beginning, especially in regulated environments.
• Communication & working style: You explain complex technical concepts in an understandable way, work pragmatically and move confidently even in dynamic project environments.
• Advantage: Experience with vLLM, SGLang or similar inference technologies, GPU clusters, sovereign cloud or private cloud environments.
• On the road: You are ready to travel and flexible to advise our customers on-site throughout Germany.
• Let's talk: You speak fluent German and English - great, then you'll find your way around Exxeta perfectly.

At Exxeta, we develop digital solutions that truly make a difference – in companies, markets, and minds. Over 1200 colleagues bring together technology, ideas, and diverse perspectives to achieve this. What drives us: curiosity, team spirit, and the ambition to create real impact. High-tech with a heartbeat.

We are a home for people who want to make a difference. Diversity and different perspectives enrich our team. Attitude, ideas, and a passion for action – that's what counts for us!

Apply now!