Forward Deployed Engineer AI Inference (Intern)

About Lyceum

Lyceum is a sovereign European AI inference provider. We run open-source models on our own GPU infrastructure, powered by 100% renewable energy, so teams can build with AI on their own terms – without giving up their data or getting locked into a single vendor. Backed by tier-1 investors, we're growing fast and our inference business is about to scale strongly.

The Role

As a Forward Deployed Engineer Intern, you own the technical side of our AI inference deals. You help customers figure out which models, GPUs and configurations fit their needs, run technical sessions with them, and work hand in hand with our commercial team to get deals closed.

This is not a pure engineering role: you'll spend a lot of time with customers, and we're looking for someone who enjoys exactly that. Much of this work is still manual today – you'll help us turn it into product.

What You'll Do

• Match customer requirements to the right models, GPUs and configurations for dedicated inference
• Run technical sessions with customers and help them make confident decisions
• Work in tandem with our commercial team to move deals forward
• Support serverless and API customizations, and help turn recurring ones into product
• Translate customer needs into clear technical specs for our engineering team
• Collect benchmarks and learnings that help us automate matching and customizations

What We're Looking For

• Studies in computer science, data science or a closely related field
• Interest in or first exposure to AI inference: LLMs, inference engines, GPUs
• Real excitement about working with customers and the commercial side – not just the technical one
• Strong communication skills: you talk confidently to customers and engineers alike
• An entrepreneurial mindset: give you an outcome, and you find a way without getting blocked
• You stay calm and constructive when your ideas are challenged
• Fluent English

Bonus Points

• Coursework or projects on inference engines or ML systems (e.g. vLLM, SGLang, TensorRT-LLM)
• Experience with GPU sizing, model serving, benchmarking or performance optimization
• A previous internship at an AI infrastructure or inference company
• Startup experience
• German

Why Join Us

• Cutting-edge work: Solve real AI inference problems with real customers from day one
• Rare mix: Combine technical and commercial work – unusual for an internship
• Real ownership: Help build what becomes our product, from GPU matching to customizations
• Founder access: Work directly with our product team and the founders
• Mission-driven team: Build sustainable, 100% renewable compute for the AI era, backed by top-tier investors