About this RoleWe are looking for an Infrastructure / Cluster Engineer to design, build, and operate the cluster infrastructure behind Gimlet's heterogeneous inference cloud. Unlike traditional cloud platforms built around a single hardware ecosystem, Gimlet's infrastructure spans multiple accelerator vendors and architectures. Infrastructure engineers play a key role in bringing new hardware platforms online, building the operational abstractions that make heterogeneous infrastructure manageable at scale, and ensuring new silicon can serve production workloads reliably from day one.
This role is highly hands-on. You will work across bare metal, Linux, Kubernetes or cluster schedulers, high-speed networking, observability, provisioning, and incident response. You will partner closely with distributed systems, runtime, compiler, and hardware teams to ensure Gimlet's infrastructure can support demanding AI workloads at production scale.
What you will work on- Design, deploy, and operate large-scale CPU, GPU, and accelerator clusters powering production AI inference.
- Build automation for provisioning, configuration, upgrades, validation, and lifecycle management.
- Design and scale provisioning systems for heterogeneous bare-metal infrastructure across multiple datacenters and hardware vendors.Operate cluster scheduling, resource allocation, isolation, quotas, and utilization systems.
- Debug complex production issues across Linux, networking, storage, drivers, firmware, and orchestration layers.
- Build and operate high-performance networking infrastructure, including RDMA-enabled environments and accelerator interconnects.
- Build observability for cluster health, capacity, performance, failures, and workload behavior.
- Improve reliability, availability, and recovery across multi-node production systems.
- Work with distributed systems and runtime teams to support low-latency, high-throughput inference workloads.
- Evaluate and integrate new hardware platforms, accelerators, networking technologies, and datacenter designs.
- Create runbooks, operational standards, and incident response practices as the fleet scales.
You may be a good fit if- Experience in infrastructure, cluster engineering, platform engineering, SRE, HPC, or distributed systems.
- Deep Linux systems experience, including debugging performance, networking, storage, processes, and kernel-level issues.
- Experience operating Kubernetes, Slurm, Nomad, or similar orchestration and scheduling systems.
- Strong automation skills using tools such as Terraform, Ansible, Helm, Python, Go, or equivalent.
- Experience with GPU or accelerator infrastructure, including drivers, firmware, CUDA/ROCm stacks, or hardware validation.
- Familiarity with high-performance networking such as InfiniBand, RoCE, high-speed Ethernet, or datacenter fabrics.
- Strong operational judgment: you know how to build systems that are observable, recoverable, and boring in production.
- Comfort working in a fast-moving startup environment with high ownership and ambiguity.
Strong candidates may also have- Experience building or operating AI inference, training, HPC, or neocloud infrastructure.
- Experience with bare-metal provisioning, PXE/iPXE, image pipelines, BIOS/firmware management, or rack bring-up.
- Experience with multi-tenant cluster isolation, quota systems, fair scheduling, or usage accounting.
- Experience debugging distributed workload performance across compute, memory, network, and storage bottlenecks.
- Experience building observability platforms using technologies such as Prometheus, OpenTelemetry, Grafana, or similar tooling.
- Familiarity with heterogeneous hardware environments across NVIDIA, AMD, Intel, ARM, or emerging accelerators.
