Forward Deployed Engineer

Role overview

Own the technical success of a production software deployment inside a customer’s AI datacenter. The role combines hands-on infrastructure engineering, troubleshooting, and performance work with a feedback loop into product development; it also contributes field expertise to evaluations and deployments with prospective customers.

Responsibilities - Serve as the primary technical owner for a customer deployment across GPU or XPU clusters, networking fabrics, schedulers, and orchestration systems. - Lead integrations, upgrades, cluster expansions, performance tuning, and root-cause investigations for production issues. - Create environment-specific automation, runbooks, and integration guides; prototype fixes and tools using Python, Bash, or APIs. - Work with customer infrastructure, SRE, and platform teams, translating operational needs into product requirements with engineering colleagues. - Share structured feedback on bugs and feature gaps, and coordinate deployment milestones with product releases. - Support technical evaluations, proof-of-concept deployments, reference architectures, and customer workshops.

Requirements - At least five years in infrastructure engineering, SRE, platform engineering, solutions architecture, or a similar hands-on role with customer exposure. - Experience operating AI datacenter clusters, including accelerators, high-speed interconnects, scheduling, and training or inference workloads. - Strong Linux, container, and API skills, plus the ability to debug distributed systems and write production-quality Python and Bash tools. - Practical experience with infrastructure platforms such as Kubernetes, Slurm, Tinkerbell, Canonical MaaS, or similar systems. - Ability to communicate clearly with engineers and senior infrastructure stakeholders; willing to work on-site and travel as needed.

Nice to have - Experience embedded with a major customer or launching early-stage products in large AI environments. - Familiarity with inference serving, GPU utilization, KV-cache behavior, or disaggregated serving. - Experience with Prometheus, Grafana, OpenTelemetry, or infrastructure analytics.

Benefits and work setup - On-site at a customer office in Cambridge, UK, with travel as needed. - Target base salary of £115,000–£135,000 per year, plus equity.