Top Tech Jobs & Startup Jobs

Design and build distributed systems for AI workloads, improve reliability and efficiency, and develop infrastructure automation tools.

Architect and build a highly available Kubernetes control plane for an AI-native cloud platform managing thousands of clusters. Develop Operators and CRDs, optimize etcd, implement multi-tenant security, enable zero-downtime upgrades and federation, and integrate with GPU orchestration systems. Lead distributed systems architecture, reliability engineering, design reviews, and team mentorship while preventing control-plane failures from affecting customer workloads.

Own end-to-end reliability for a customer-facing GPU cloud serving workloads across 100–10,000 GPUs. Responsibilities include operating Kubernetes GPU clusters, managing multi-tenant isolation, automating bare-metal provisioning, defining SLIs/SLOs/SLAs, handling incidents and customer communications, building observability with Prometheus and Grafana, implementing infrastructure as code and GitOps, and developing automated remediation and self-service operational tooling.

Lead identification, application, and management of US federal, state, and local tax abatement, exemption, and incentive programs (85%). Support US tax advisory, compliance, and liaison with tax authorities (15%), model financial impact, maintain centralized tracking, manage renewals, and coordinate with stakeholders and external advisors to ensure compliance and optimize tax savings.

Owns architecture, deployment, operations, reliability, and multi-tenant scheduling policy for production Slurm HPC GPU clusters across bare-metal and virtualized environments. Leads Slinky Kubernetes integration, elastic capacity sharing, containerized workloads, GPU and fabric health monitoring, infrastructure automation, observability, accounting, and billing integration. Provides customer onboarding and escalation support, writes operational documentation, leads incident response, and mentors platform engineers.

Deploy, operate, and optimize high-performance parallel/distributed storage systems for AI training and inference; implement multi-tenant isolation and GPU Direct data paths; instrument telemetry for storage-fault prediction; convert incidents into automated runbook-as-code; plan capacity, firmware, migrations, and DR for GPU clusters.

Front-line SRE covering 8AM–8PM PST monitoring GPU clusters, networking, storage, and sensors. Execute runbooks, perform hardware triage and physical DC tasks (rack, cable, swap), collect diagnostics for escalation, manage tickets (ServiceNow/Jira), update runbooks, and tag incidents to train the AIOps platform.

Build global, multi-region cloud control-plane services for GPU infrastructure, including tenant management, quotas, metering, billing, ticketing, RBAC, customer portals, and APIs. Design cross-region consistency, automate customer onboarding, ensure billing auditability, develop operational recovery tooling, and support region-aware scheduling and placement.

Design, implement, and maintain CI/CD and MLOps pipelines, provision and scale cloud-native AI infrastructure (Kubernetes, GPU clusters), enforce IaC practices, implement observability and security frameworks, lead incident response and cross-functional platform engineering to ensure high availability and compliance.

Lead end-to-end architecture of AI data center networks, DCI, and global backbone for large GPU clusters. Design underlay and overlay networks, IP/VLAN/VXLAN planning, congestion control (PFC/ECN/INT), and optical transport. Produce HLD/LLD, topology diagrams, standards, and SOPs; engage vendors and drive risk mitigation and network improvements.