Senior AI Scientist: Efficient LLM Inference & Model Optimization

Nebius Group; Greater London; On-site; GBP 120,000 - 160,000; Full time

Benefits offered by this job

Competitive compensation

Career growth and learning

Flexibility and ownership

Collaborative and innovative culture

Impactful AI projects

International environment

Job summary

Token Factory is seeking a Senior Applied Scientist to turn frontier inference bottlenecks into credible research and production-ready solutions. You will design rigorous experiments, write strong code, and collaborate with engineers to ship results.

You will own well-scoped research projects, publish or prepare high-quality technical work, and prototype components that engineers can productionize. A strong research background and coding ability are essential for impact.

Qualifications

• PhD in CS/ML or closely related field.
• Strong publication record or equivalent in ML/efficient inference.
• Hands-on coding in Python and PyTorch; move from idea to experiment quickly.
• Deep understanding of LLMs, VLMs, transformer inference, and production-serving tradeoffs.
• Strong experimental design skills with ablations, baselines, metrics and analysis.
• Excellent written and verbal communication.

Responsibilities

• Own focused research projects from hypothesis through production handoff.
• Publish internal reports, blogs, or papers when credible externally.
• Collaborate with MLEs to productionize research prototypes.
• Define research programs in efficient LLM and VLM inference with measurable impact.
• Build prototypes in PyTorch, Triton, CUDA-adjacent tooling; work with engineers to productionize.
• Design evaluation methodology covering quality, latency, memory, and cost per token.
• Publish open-source artifacts and technical reports for external credibility.
• Collaborate with cross-functional teams to choose high-leverage research bets.
• Mentor engineers and scientists on experimental design and tradeoffs.

Skills

Python

PyTorch

LLM inference

Experimental design

Publication track record

Technical writing

Communication

Education

PhD in CS / ML / ML systems

Tools

CUDA

Triton

NVIDIA TensorRT

Model compression / quantization tooling

Job description

Token Factory is seeking a Senior Applied Scientist to turn frontier inference bottlenecks into credible research and production-ready solutions. You will design rigorous experiments, write strong code, and collaborate with engineers to ship results.

You will own well-scoped research projects, publish or prepare high-quality technical work, and prototype components that engineers can productionize. A strong research background and coding ability are essential for impact.