SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Baseten is seeking a Delivery Director for Capacity Programs to own end-to-end delivery of large-scale GPU and compute infrastructure. This role sits at the intersection of infrastructure engineering, capacity planning, vendor management, and customer delivery—translating ambitious infrastructure roadmaps into predictable, on-time execution.
Key responsibilities include:
- Own delivery of on-premises data center builds: colocation expansions, power/cooling readiness, rack-and-stack, network fabric bring-up, and hardware acceptance testing, coordinating across colo providers, network engineering, hardware ops, and vendor teams.
- Drive neo cloud delivery programs: manage capacity delivery from GPU cloud and neo cloud partners, including contract milestones, capacity ramps, SLAs, and go-live readiness.
- Build and maintain master delivery schedules across concurrent, multi-site, multi-vendor programs, integrating power/shell timelines, hardware lead times, logistics, and software/platform readiness into a single critical path.
- Manage cross-functional stakeholders (engineering, data center operations, procurement, finance, legal, executive leadership), translating technical dependencies into clear program status, risks, and decisions.
- Own vendor and partner relationships for infrastructure delivery, including neo cloud providers, colocation partners, hardware OEMs, and logistics/installation contractors; hold partners accountable to contractual commitments.
- Run risk and dependency management: identify critical path risks (power availability, hardware lead times, permitting, supply chain), drive mitigation, and escalate proactively.
- Establish delivery governance: standardized program frameworks, milestone definitions, RAID logs, capacity tracking, and reporting cadences (weekly ops reviews, executive/board-level updates).
- Drive capacity planning alignment with demand/capacity planning teams to ensure delivery timelines match business commitments.
- Lead post-mortems and continuous improvement after major builds or delivery milestones.
Required qualifications: 8+ years in technical program management, delivery management, or infrastructure program leadership, with at least 3+ years managing large-scale physical infrastructure or data center delivery programs. Demonstrated experience delivering on-prem data center builds from planning through go-live. Direct experience with GPU cloud/neo cloud capacity delivery, understanding of GPU cluster architectures (InfiniBand/RoCE fabrics, rack power density, liquid cooling). Track record managing complex, multi-vendor, multi-workstream programs with hard external deadlines. Strong executive communication skills and proven vendor/partner management experience.
Preferred: hyperscaler, neo cloud provider, colocation, or AI infrastructure company experience; familiarity with GPU hardware lifecycles (NVIDIA H100/H200/GB200); PMP/PgMP certification; experience building delivery/program management functions from the ground up.
