AI Cluster Technical Program Manager – Validation, Debug & Agentic AI

Advanced Micro Devices, Inc•Austin, TX

43d•Onsite

About The Position

We are seeking a Technical Program Manager to lead execution of AI cluster engineering programs with deep focus on GPU platforms, rack-level solutions, and AI Cluster validation. This role is responsible for driving end-to-end delivery from GPU + server integration through rack bring-up, scale testing, failure analysis, and system debug closure, ensuring platform readiness for hyperscale and enterprise AI deployments. This role operates at the intersection of hardware, firmware, networking, and scale-test execution, and requires strong technical depth combined with disciplined program execution. You are a hands-on TPM who thrives in complex, fast-moving ecosystems, and can connect deep technical details to crisp program plans, executive reporting, and customer outcomes. You are comfortable driving execution in bring-up and EVT/DVT/PVT working closely with engineers to root-cause issues, unblock debug, and make data-driven tradeoffs to keep programs moving. You bring urgency, ownership, and clarity to ambiguous problem spaces and can communicate effectively from lab floor to executive review.

Requirements

Experience leading complex hardware or AI infrastructure programs with ownership across bring-up, validation, and deployment phases.
Strong technical understanding of GPU-based AI systems, rack architectures, and datacenter infrastructure.
Proven ability to manage ambiguity, drive debug execution, and lead cross-functional teams without direct authority.
Strong written and verbal communication skills, including executive-level status reporting.
Proficiency with program management and execution tools (Jira, Confluence, dashboards, Excel/PowerPoint).
Bachelor’s or master’s degree systems, EE, CS, or related engineering discipline.
PMP, Scrum Master, or equivalent program management training.

Nice To Haves

Hands-on experience with GPU cluster scale testing, system stress, or performance validation.
Familiarity with rack-level bring-up, power/cooling constraints, networking, and failure modes at scale.
Experience working through hardware/firmware debug cycles in pre-production or customer-facing environments.
Experience managing fleet-scale validation and deployment of AI, cloud, HPC, or hyperscale infrastructure.
Strong understanding of infrastructure observability, telemetry pipelines, log analytics, and incident management frameworks.

Responsibilities

Define, plan, and drive program plans for AI infrastructure systems validation and readiness, including server integration, rack bring-up, and cluster-scale deployment readiness.
Create and maintain core PM artifacts: schedules, dependency maps, resource forecasts, risk/issue logs, and program dashboards/status reports.
Identify and drive mitigation plans for issues/risks, including cross-team escalations and corrective actions across multiple engineering areas.
Drive regular execution reviews with engineering teams and provide concise, data-driven updates to senior leadership.
Own program execution for GPU-based AI platforms, spanning system bring-up, qualification, scale readiness, and deployment validation across server, rack, and cluster levels.
Drive alignment across GPU, CPU, firmware, BIOS/BMC, and system teams to ensure readiness for scale testing and customer workloads.
Track platform issues, and debug dependencies; ensure risks are clearly documented, owned, and mitigated.
Own program planning and execution for multi-node and multi-rack scale testing, including test strategy, scheduling, coverage tracking, and readiness gates.
Lead end-to-end delivery of rack-level AI solutions, including compute trays, switch trays, cabling, power, cooling, and management infrastructure.
Ensure rack bring-up plans are executable, resourced, and gated with clear entry/exit criteria across EVT, DVT, and scale phases.
Drive coordination across lab operations, infrastructure, and engineering teams to unblock rack access, power, networking, and test readiness.
Partner with scale, performance, and automation teams to ensure workloads, stress tests, and regressions plans are ready before hardware arrives.
Act as the execution lead for platform debug, coordinating across engineering teams to ensure fast triage, root-cause analysis, and resolution of system-level issues.
Track high-impact failures (GPU, HSIO, FW, rack, network) through debug forums ensuring clear ownership and closure plans.
Balance debug depth vs. program timelines, escalating tradeoffs when needed and ensuring leadership has a clear view of risk and impact.
Lead end-to-end system rack and cluster-level debug activities during validation, deployment, and fleet operations, driving rapid fault isolation and root-cause analysis across hardware, firmware, networking, and software domains.
Own incident management for critical deployment and validation issues, coordinating cross-functional war rooms and executive communications to ensure timely resolution.
Define key operational metrics such as Mean Time to Detect (MTTD), Mean Time to Mitigate (MTTM), Mean Time to Resolution (MTTR), incident recurrence rates, and fleet health indicators.
Drive post-incident reviews (PIRs), root-cause analysis (RCA), corrective actions, and long-term preventive measures to reduce Sev1/Sev2 incident recurrence and improve fleet reliability and deployment readiness.
Champion Agentic AI and AIOps solutions for automated incident triage, log analysis, root-cause identification, and operational workflow automation at scale.
Identify opportunities to leverage Agentic AI capabilities across validation and debug workflows.
Drive development and rollout of AI agents for: Automated triage, Log analysis, Root-cause identification, Knowledge retrieval, Test orchestration, Incident management.
Partner with data engineering teams to integrate AI-powered operational tools.
Establish success metrics and measurable business impact for agent-based automation initiatives.
Champion AI-first operational workflows across engineering organizations.