NLP Performance Engineer
G-Research · London, London, GB
Quant Financeonsite
At a glance
Highlights
- Own large-scale LLM inference performance on cutting-edge compute infrastructure.
- Design reference solutions that improve efficiency and reliability of NLP workloads.
- Collaborate with research and platform teams to evolve the compute stack.
What they offer
- Highly competitive compensation plus annual discretionary bonus
- 35 days' annual leave
- 9% company pension contributions
- Lunch provided and dedicated barista bar
- Comprehensive healthcare and life assurance
- Informal dress code and excellent work/life balance
- Cycle-to-work scheme
- Monthly company events
Ideal candidate
A performance-obsessed engineer with a scientific, evidence-led mindset who loves digging into bottlenecks and partnering with researchers to unlock faster, more efficient AI systems.
Requirements
- Bachelor's, Master's or PhD in computer science, or equivalent experience
- Proven experience profiling, benchmarking and optimising large-scale LLM inference workloads
- Scientific, evidence-led approach to performance optimisation
- Deep understanding of transformer inference including prefill vs decode and KV-cache
- Hands-on experience with LLM serving frameworks such as vLLM, SGLang, TensorRT-LLM or TGI
- Experience with inference optimisation techniques including quantisation and speculative decoding
- Strong software engineering skills including Python, CUDA and reliable ML systems
- Strong communication skills with ability to collaborate across teams
Lab profile
About G-Research
Quantitative finance research firm bringing scientific clarity to financial complexity.
Known for
- Quantitative finance research
- Research and engineering collaboration
- Platform development
Key facts
- Headquartered in London
First seen · Last seen