GPU Infrastructure Lead - Banking & Finance

Hamilton Barnes; Greater London; Hybrid; GBP 135,000 - 165,000; Full time

Benefits offered by this job

Hybrid work from London or remote

Early-stage equity

Job summary

Hamilton Barnes is seeking a Lead Infrastructure Validation Engineer to own the technical strategy and hands-on execution. The role offers hybrid work in London or fully remote, with a focus on certifying large GPU clusters to meet exchange-traded contract standards.

You will author validation code and rack-level runbooks, set engineering standards for compute and network topology, and lead a growing team as the business scales in fintech infrastructure.

Qualifications

• Proven hands-on GPU cluster deployment at scale, with leadership or technical authority.
• Deep NVIDIA stack familiarity including HGX, NVLink/NVSwitch, CUDA debugging.
• Experience with infrastructure-as-code, monitoring and automated testing for hardware.
• Ability to communicate clearly with traders, lawyers, providers and regulators.

Responsibilities

• Design and run cluster validation and certification, including benchmarks and interconnect testing
• Build deployment automation and health checks for scalable certification
• Set standards for compute; node configurations, network topologies, storage and cooling envelopes
• Collaborate with capacity providers through onboarding and testing
• Influence contract specs, index methodology and product roadmap with field learnings
• Hire and lead the infrastructure engineering team as the business scales

Skills

GPU clustering

NVIDIA stack

NCCL performance

Infrastructure-as-code

Monitoring & automation

Tools

HGX platforms

NVLink/NVSwitch

CUDA debugging

Monitoring stack

Job description

Are you looking for an exciting new opportunity?

Join a pioneering fintech scaling the market infrastructure that transforms GPU compute into a tradeable,financeable asset class.Bridging hardware and financial market structure across price indices,physical capacity trading,and regulated derivatives,the business delivers physical validation standards for high-performance AI infrastructure.

The organization is seeking a Lead Infrastructure Validation Engineer to own technical strategy and hands-on execution.Operating hybrid in London or fully remote,the role involves establishing certification standards for large-scale GPU clusters,authoring validation code and rack-level runbooks,and ensuring physical compute infrastructure meets exchange-traded contract and collateral specifications.

Responsibilities

• Design and run cluster validation and certification, covering performance benchmarks, interconnect testing (NCCL, InfiniBand/RoCE), thermal and power verification, and SLA availability monitoring
• Build automation for deployment, health checks and continuous testing so certification scales without headcount scaling alongside it
• Set the technical standards for deliverable compute, including node configurations, network topologies, storage and cooling envelopes
• Work directly with capacity providers (neoclouds and data centre operators) through onboarding, from site walkthroughs to acceptance testing
• Feed ground-level learnings back into contract specifications, index methodology and the product roadmap
• Hire, build and lead the infrastructure engineering team as the business scales

Skills/Must have

• Experience: Proven hands-on GPU cluster deployment at both ends of the scale, from a handful of nodes to hundreds or thousands of GPUs, with team leadership or de facto technical authority (the person everyone routes hard problems to)
• Core Tech/Domain: Deep NVIDIA stack familiarity (HGX platforms, NVLink/NVSwitch, CUDA-level debugging, NCCL performance tuning), across both air-cooled and liquid-cooled estates
• Methodology/Protocols: Infrastructure-as-code, monitoring and alerting stacks, and automated acceptance and regression testing for hardware. Real production scar tissue: cabling and topology errors, firmware mismatches, failing optics, thermal throttling, and the gap between passing burn-in and surviving production
• Soft Skills: Ability to set direction and bring people with you, and to communicate clearly with non-infrastructure stakeholders such as traders, lawyers, capacity providers and regulators

Desirable Skills

• Direct liquid cooling (DLC) deployment experience
• Operator-side background at a neocloud, hyperscaler, HPC centre, or a crypto-mining-to-AI conversion
• Exposure to benchmarking frameworks (MLPerf, custom acceptance suites) or SRE practice at scale
• Interest in, or exposure to, financial markets and how physical commodities are standardised and traded

Benefits

• Hybrid working from London, or fully remote
• Early-stage equity and wider benefits: to be confirmed with the client before release

Salary

• £150k