Senior AI Compute Architect – GPU Infrastructure

Kraken; United States; Hybrid; GBP 90,000 - 130,000; Full time

Job summary

Kraken is building a dedicated AI Compute and Infrastructure team to power next‑gen model training, inference, evaluation, and experimentation. This role owns GPU clusters, drivers, runtimes, and workload isolation to enable Kraken's AI workloads with speed and cost discipline.

You will collaborate with ML researchers, platform engineers, security teams, and product teams to ensure fast, dependable, and production‑grade compute infrastructure that scales with Kraken's AI ambitions.

Qualifications

• 5+ years of infrastructure engineering experience, with GPU compute or ML infra.
• Hands-on experience operating GPU clusters in production.
• Strong systems engineering: Linux, networking, storage, containers, Kubernetes.
• Experience with ML serving frameworks (vLLM/Triton/TensorRT/TorchServe/KServe).
• Proficiency in Python for automation and tooling.
• Understanding of performance vs cost and reliability tradeoffs.
• Experience building observable systems with metrics, logs, and alerts.

Responsibilities

• Own and operate GPU/accelerator clusters for training, inference, and experimentation.
• Design infra to allow model runs on GPUs locally to reduce external dependencies.
• Build and improve scheduling, orchestration, and resource utilization systems.
• Optimize inference pipelines for latency, throughput, memory, and cost.
• Collaborate with ML engineers to remove bottlenecks in training and deployment workflows.
• Build observability for GPU utilization, latency, and spend.
• Drive reliability, incident response, and runbooks for AI compute infra.
• Evaluate new hardware, accelerators, runtimes, and serving frameworks.
• Create tooling to make GPU usage transparent for internal teams.
• Contribute to long-term architecture balancing performance and production safety.

Skills

GPU compute

ML infrastructure

Python

Linux

Kubernetes

Performance optimization

Tools

vLLM

Triton Inference Server

TensorRT

TorchServe

KServe

Ray Serve

Job description

Kraken is building a dedicated AI Compute and Infrastructure team to power next‑gen model training, inference, evaluation, and experimentation. This role owns GPU clusters, drivers, runtimes, and workload isolation to enable Kraken's AI workloads with speed and cost discipline.

You will collaborate with ML researchers, platform engineers, security teams, and product teams to ensure fast, dependable, and production‑grade compute infrastructure that scales with Kraken's AI ambitions.