Senior Technical Program Manager, DGX Cloud Software Products and Services

Description

The company runs a cloud product that supports AI work for its own research and products. A central performance program guides every cloud service provider (CSP) and the company Cloud Partner (NCP) toward strong results, and measures success by total workload cost of ownership rather than GPU cost per hour.

The program gives a single, consistent measure of what the company's hardware and software deliver at scale. It sets the reference point for cloud partners and corrects inaccurate GPU comparisons published by analysts.

The company is hiring a Senior Technical Program Manager to run the day-to-day work of this program. You will turn priorities into plans for capacity, validation, debugging, and releases that line up across CSPs, NCPs, and internal engineering teams.

Your workstreams cover Slurm and Kubernetes benchmarking for training and inference, debugging across partners, release and QA coordination, PLC and OSRB alignment, and visibility into program status. The program is moving into inference and agentic AI, and you will apply the same methods to those stacks and new partner work.

You will work directly with the program lead and the Product lead to keep execution in line with the roadmap. The role needs strong technical skill, self-direction, and the ability to work with engineering, product, and partner teams at every level. The ideal candidate understands cloud infrastructure and inference systems well, asks sharp questions, drives investigations to a conclusion, and reduces technical problems to short, clear next steps.

Responsibilities

• Track demand and allocation across platforms, GPU counts, topologies, and workloads, and arrange capacity reservations, access, readiness checks, and debug windows with cluster owners.
• Follow cross-CSP performance investigations with engineering, solutions architects, and partner teams, and build playbooks and pre-flight checks for problems that keep recurring.
• Keep a coordinated plan for achievements and test coverage across recipes, frameworks, results tooling, and QA. Chase merge requests, reviews, dependencies, and regressions that block progress, and write release readiness summaries for each program increment.
• Connect software roadmap delivery to full-stack readiness, covering infrastructure, bring-up, validation, and the operational support that follows.
• Set common processes for intake, debug handoffs, and release evidence, and extend them from training into inference, Kubernetes, and agentic benchmarking as priorities shift.
• Set up a regular engineering and product cadence for PLC achievements and monthly releases.
• Find cross-functional dependencies, reduce risk, and drive complex technical and program issues to resolution across partner, engineering, and product teams.
• Build and maintain metrics for program efficiency and delivery health. Gather and analyze data to inform planning and decisions, and report progress clearly to senior leadership.

Requirements

• 8+ years of technical program management experience.
• A Bachelor's degree in Computer Science, Electrical Engineering, AI, or a related field, or equivalent experience, is required.
• You have shown you can run programs across multiple time zones and organizations.
• Working knowledge of inference systems, cloud platforms, and distributed computing, enough to ask relevant technical questions and read engineering updates directly.
• You have run complex, multi-stakeholder programs in fast-moving software teams, from kickoff through delivery to partners.
• You have a background in software release management, QA coordination, and platform validation cycles. Experience with compliance or review routines such as open-source review, security review, or PLC/OSRB is preferred.
• You write clear, executive-ready status updates and blocking-issue memos that put the key point first.
• You have used large-scale Agile and reporting tools such as Jira, NVBugs, or equivalent, and you are comfortable designing your own tracking structure when none exists.
• Ways to stand out: a background in machine learning, deep learning, and AI applications, with familiarity with AI benchmarking tools, evaluation frameworks, or performance scorecards for large GPU workloads.
• Ways to stand out: experience with LLM inference frameworks such as vLLM, SGLang, or TensorRT-LLM, or with inference optimization methods such as speculative decoding, quantization, disaggregated prefill/decode, or multi-token prediction.
• Ways to stand out: prior work supporting CSP or NCP partner programs, including platform validations, cluster capacity requests, or performance bring-up across organizations.
• Ways to stand out: Kubernetes or cloud-native workload experience, including how container orchestration affects GPU placement, scheduling, and repeatable workloads.
• Ways to stand out: experience with open-source release procedures, including contributor licensing, OSRB-style assessments, and ecosystem readiness, for publishing internal tools in a compliant, organized way.

Conditions

Applications will be accepted at least until November 6, 2026.

This is an existing vacancy.

The company uses AI tools in its recruiting process.

The company is an equal opportunity employer and does not discriminate on the basis of any protected characteristic.

Benefits

• Competitive salary and a comprehensive benefits package. Details are on the company's benefits page.