Senior Technical Program Manager, Deep Learning Frameworks

Details

Deep Learning Software is looking for a Technical Program Manager to lead software programs for deep learning training and inference frameworks. The role supports NVIDIA's AI researchers and software engineers and involves managing software programs in a highly matrixed environment with an active and evolving roadmap. The TPM will work with senior management and partners across the company, coordinate software, hardware, and infrastructure teams, and develop standardized planning, reporting, execution practices, and metrics.

Responsibilities

• Collaborate with hardware, software, and model program managers, product managers, and engineering teams to deliver deep learning framework programs on existing and new hardware.
• Engage cross-company hardware engineering, software engineering, product, QA, and compliance teams to align on release scope, milestones, risk management, and dependencies.
• Lead program management activities including planning, forecasting, documentation, scheduling, meetings, prioritization, dependency management, reporting, and resolution of critical and blocking issues.
• Develop and implement metrics to measure program effectiveness and identify areas for improvement.
• Collect and analyze data to support planning and data-driven decisions.
• Define and implement standard processes for open-source contribution and release management within the NVIDIA AI/ML ecosystem.
• Report overall program status and provide insights and recommendations to senior management.
• Coordinate with multifunctional leads to drive organizational alignment and efficiency and streamline processes.

Requirements

• Postgraduate degree in Computer Science, Artificial Intelligence, Mathematics, or equivalent experience.
• 10+ years of software program management experience, including a proven record of leading global projects across multiple time zones in fast-paced software development environments.
• Experience delivering large software programs spanning multiple layers of the software stack.
• Ability to think strategically and tactically, build consensus, and moderate effective engagements with engineering and product teams.
• Excellent communication, technical presentation, organizational, and attention-to-detail skills.
• Ability to multitask in a dynamic environment with shifting priorities and changing requirements.
• Experience using project management tools such as Jira, Aha!, and Confluence.
• Experience with distributed version control systems such as Git.

Preferred Qualifications

• Experience with deep learning frameworks such as PyTorch and JAX.
• Experience with ML compilers such as XLA and Triton.
• Experience with GPU technology and open-source development.
• Prior production software development, release management, or DevOps experience.
• A consistent record of driving process improvements and measuring efficiency.
• Exposure to the NVIDIA GPU programming and software stack, including CUDA Toolkit, cuDNN, TensorRT, and NCCL.
• Engineering background.

Compensation and Benefits

• Base salary range for Level 4: USD 168,000–258,750 per year.
• Base salary range for Level 5: USD 200,000–322,000 per year.
• Eligibility for equity and benefits.