Kubernetes Engineer
Tensormesh · San Francisco Bay Area · 1 wk ago
EngineeringFull-time
About the role
We are seeking a Kubernetes Engineer to build the core orchestration logic of the Tensormesh platform. Unlike a standard DevOps role, this position involves deep software development within the Kubernetes ecosystem. You will extend Kubernetes capabilities by writing custom operators and controllers that manage complex AI inference workloads, ensuring high availability and seamless auto-scaling.
What You’ll Do
- Develop Custom Operators: Design and implement Kubernetes Custom Resource Definitions (CRDs) and Controllers (using Golang/Kubebuilder) to manage the lifecycle of Tensormesh products.
- Traffic Management: Architect and build high-performance ingress and load-balancing systems capable of handling high-throughput inference requests.
- Resilience & Scaling: Develop logic for intelligent auto-scaling (HPA/VPA) based on GPU metrics and ensure High Availability (HA) for critical components.
- K8s Optimization: Tune the scheduler and resource management configurations to maximize GPU utilization for inference tasks.
Ideal candidate credentials
- 2-10 years of software engineering experience, with a focus on distributed systems or container orchestration.
- Strong proficiency in Go (Golang); experience with the Kubernetes client-go library and Kubebuilder/Operator SDK is highly preferred.
- Deep understanding of Kubernetes internals (API machinery, Controller runtime, Networking, CNI).
- Experience with service meshes (Istio, Linkerd) or ingress controllers (Nginx, Traefik) is a plus.
- Understanding of distributed consensus and state management.