Staff Technical Program Manager, Cluster Software Platform
Sunnyvale, CA · Remotefull time
Remote
senior
H1B: yes
Cerebras Staff Technical Program Manager at Cerebras leading multi-team infrastructure and platform programs for Cluster Software. Responsible for release management, fleet-wide rollouts, and program operating model design.
Responsibilities
Cluster Software is the platform that turns Cerebras wafer-scale systems into reliable, secure inference and training capacity, from a single CS system to multi-site deployments of tens of megawatts. It spans the cluster control plane, workload orchestration, monitoring and observability, deployment and data center bring-up automation, the fleet operations platform, test and release qualification, and security. It now also extends to disaggregated inference, where GPU platforms from multiple vendors run alongside Cerebras systems, and to clusters our customers and partners operate on their own premises.
As a Staff+ TPM for the Cluster Software Platform, you run key technical programs related to automation of datacenter bringup/daily operations, new client bringup, multi-tenancy, etc. Drive cross-team decisions related architecture trade-offs, release boundaries, staffing trade-offs, and acceptance criteria.
Own the integrated Cluster Software roadmap for one or more technical foundations across control plane, orchestration, monitoring, deployment, operations platform, test and release, and security
Drive production cluster reliability –related postmortems and monitoring dashboards
Shape quarterly objectives with engineering and product leaders, make dependencies and staffing conflicts explicit, and drive leadership decisions when priorities compete.
Partner with architects on interface contracts and compatibility, so new capabilities land without destabilizing production.
Drive the operations platform that replaces manual fleet work with governed, auditable APIs and site agents supporting a zero-operator-access model where customers require it.
Own the release strategy and train: the combined cluster release today and the move to independently versioned, independently qualified components. Set scope-lock, qualification, soak, rollout and freeze policies, and track GA through verified fleet adoption.
Ensure every production change has go/no-go criteria, health validation, and a tested rollback plan.
Lead Cluster Software readiness for new programs and/or client requests for Cerebras -operated and on-premises clusters
Design the program operating model for the organization: planning cadence, decision logs, release and readiness reviews, and executive reporting.
Mentor and set standards for other TPMs; build reusable templates, metrics and processes that scale as the fleet grows.
Qualification
Working knowledge of KubernetesExecutive communicationWhy Join Cerebras
Required
10+ years in technical program management, engineering leadership or systems engineering, including 6+ years as a TPM leading multi-team infrastructure or platform programs.
Experience with data center bring-up automation, infrastructure-as-code, GitOps (Argo CD, Helm) and network or firewall automation.
Working knowledge of Kubernetes, Linux, networking, CI/CD and release engineering; able to review designs and challenge technical plans.
Experience running release trains and fleet-wide rollouts for production systems with customer SLAs.
Executive communication: clear written plans, decision memos and risk reporting to senior leadership.
Preferred
Experience with AI or HPC accelerator fleets (Cerebras, GPU, TPU, Trainium or similar), including disaggregated or heterogeneous inference.
Experience with observability platforms (Prometheus, Grafana, Loki or equivalent) and SLA reporting.
Experience productizing a platform for customer-operated or on-premises deployments.
Experience building and mentoring a TPM function.
Why Join Cerebras
People who are serious about software make their own hardware. At Cerebras, we have built a breakthrough architecture that is unlocking new opportunities for the AI industry. With dozens of model releases and rapid growth, we’ve reached an inflection point in our business. Members of our team tell us there are five main reasons they joined Cerebras:
Build a breakthrough AI platform beyond the constraints of the GPU.
Publish and open source their cutting-edge AI research.