Member of Engineering (Inference Infrastructure)

Role overview This role sits on the compute team, partnering closely with the inference team to push the throughput and latency of large-model inference workloads. The work blends GPU workload scheduling, serving optimization, and the systems engineering needed to keep high-throughput data planes healthy under pressure. It is a deeply technical position aimed at compounding research and deployment velocity.

Responsibilities - Build and refine GPU workload scheduling and inference serving systems used for evals and reinforcement learning - Improve inference throughput and latency in collaboration with the inference team - Design and operate distributed systems, schedulers, control planes, or high-throughput data planes - Work hands-on with Kubernetes internals, including controllers, informers, and operators, rather than treating it as a deployment target - Champion observability and debuggability so production issues can be navigated quickly - Continuously raise the bar for systems performance and architectural velocity

Requirements - Strong programming skills in Go or a comparable systems language - Systems engineering background across distributed systems, schedulers, control planes, or high-throughput data planes - Production experience with Kubernetes internals, including controllers, informers, and operators - A bias toward observability and debuggability when designing production systems - Ability to thrive in a fast-paced, mission-driven environment with low-ego, collaborative teammates

Nice to have - Experience building systems that serve large-scale inference requests

Benefits and work setup - Fully remote role with flexible hours, based in the EMEA region - Roughly 37 days per year of vacation and holidays - Health insurance allowance for the employee and dependents - 16 weeks of flexible, full-pay parental leave - Well-being, always-be-learning, and home office allowances - Company-provided equipment - Regular in-person team gatherings, including monthly meetups in Paris and an annual longer off-site