Forward Deployed Engineer AI Inference

Lyceum; Berlin; On-site; EUR 110,000 - 150,000; Full time

Job summary

Lyceum in Berlin seeks an experienced senior technical owner to drive the technical delivery of our largest inference deals. You will partner with CTOs and ML leads to design setups, size GPUs, and optimize latency and cost.

You will co-create a playbook with the Sales Engineering lead and align engineering with commercial goals. You will own end-to-end deals, run benchmarks, and translate results into clear recommendations.

Qualifications

• Degree in computer science, data science or closely related field.
• 4+ years in a customer-facing technical role such as solutions or sales engineering or ML engineering with customers.
• Hands-on experience with LLM inference and model serving and GPU sizing.
• Track record owning the technical side of complex B2B deals with enterprise customers.
• Able to explain trade-offs clearly to CTOs and engineers.
• Entrepreneurial mindset and fluent English.

Responsibilities

• Own the technical side end-to-end for the largest and most complex inference deals from discovery to go-live.
• Be the senior technical counterpart for customers' CTOs and ML leads on architecture, latency and cost trade-offs.
• Design inference setups: model choice, GPU type/count, parallelism, quantization, engine and config.
• Lead benchmarks and proofs of concept and translate results into recommendations.
• Scope customer customisations with the engineering team and decide what becomes product.
• Build playbooks and tooling to match workloads to GPUs and feed roadmap insights.
• Collaborate with the commercial team on pricing, proposals and technical contract elements.

Skills

Customer-facing tech role

LLM inference

GPU sizing

CTO communication

Entrepreneurial mindset

Fluent English

Education

Degree in CS/Data science or related field

Tools

vLLM

SGLang

TensorRT-LLM

Triton

Job description

The Role

You will own the technical side of our largest and most complex inference deals, from first discovery call to go-live. You're the trusted technical counterpart for our customers' CTOs and ML leads, you design the setups that run their models, and you work hand in hand with our commercial team to get deals closed.

You'll be one of two senior technical owners of our inference deals. Together with our Inference Sales Engineering lead, you'll also turn what we learn in the field into a repeatable playbook and product.

What You'll Do

• Own the technical side of our largest and most complex inference deals end to end, from discovery to go-live

• Be the senior technical counterpart for customers' CTOs and ML leads on architecture, sizing, SLAs, latency/throughput trade-offs and cost

• Design dedicated inference setups: model choice, GPU type and count, parallelism, quantization, inference engine and configuration

• Lead benchmarks and proofs of concept, and turn the results into clear recommendations

• Scope customer customizations with our engineering team and decide what becomes product

• Build the playbook and tooling for matching workloads to GPUs, and feed our roadmap with what you learn in the field

• Work in tandem with our commercial team on pricing, proposals and the technical parts of contracts

What We’re Looking For

• A degree in computer science, data science or a closely related field

• 4+ years in a customer-facing technical role, e.g. solutions or sales engineer, forward deployed engineer, ML engineer working closely with customers, or technical consultant

• Hands-on experience with LLM inference and model serving (e.g. vLLM, SGLang, TensorRT-LLM, Triton) and GPU sizing

• A track record of owning the technical side of complex B2B deals, ideally with enterprise customers

• You're credible with customer CTOs and engineers alike, and you explain trade-offs clearly

• Entrepreneurial mindset: give you an outcome, and you find a way without getting blocked

• Fluent English

Bonus Points

• Experience at an inference provider, GPU cloud or AI infrastructure company

• Benchmarking and performance optimization: throughput, latency, cost per token

• Experience with quantization, parallelism strategies, KV-cache and batching

• Startup experience

• German

Why us?

• Outstanding team: Work with some of the best engineers in the world, coming from hedge funds, big tech, AI startups and top universities

• Once in a lifetime opportunity: Early-stage company in the fastest-growing market in the world

• Ownership: Shape how European AI companies access GPU compute

• European mission: Build sovereign, GDPR-compliant AI infrastructure for the next generation of deep-tech