Technical Program Manager, Model Performance
Zero-to-one Technical Program Manager for Baseten's Model Performance team, building planning structures and driving execution of inference optimization and model release programs across cross-functional engineering teams.
- Compensation
- $165K–$330K / yr
- Location
- San Francisco
- Work setup
- Hybrid
- Seniority
- Not specified
- Employment
- Not specified
- Experience
- Not specified
- Category
- Project management
- Posted
Skills
Must have
- Deep technical program management experience at an organization of similar or greater complexity
- Experience program-managing model performance or inference optimization work (vLLM, TensorRT-LLM, SGLang, NVIDIA Dynamo)
- Comfort with ambiguity and zero-to-one program building
- Ability to influence without authority across engineers, managers, and leadership
- Excellent written and verbal communication for making technical programs legible to any audience
- High agency decision making with ownership and accountability
Nice to have
- Experience coordinating model release programs including day-zero launches for large-scale models
- Prior hands-on software engineering background
- Deep learning performance optimization background in training or inference
Full description
ABOUT BASETEN
Baseten powers mission-critical inference for the world's most dynamic AI companies, like Cursor, Notion, OpenEvidence, Abridge, Clay, Gamma, and Writer. By uniting applied AI research, flexible infrastructure, and seamless developer tooling, we enable companies operating at the frontier of AI to bring cutting-edge models into production. We're growing quickly and recently raised our $1.5B Series F, led by Altimeter Capital, Conviction Partners, and Spark Capital. Join us and help build the platform engineers turn to ship AI products.
THE ROLE
The Model Performance organization at Baseten is looking to hire our first Technical Program Manager. This is a zero-to-one role in a team that is responsible for building the core algorithms and methods that power Baseten’s high performance inference stack. You won't inherit an existing program framework, you'll build one from the ground up: the planning structure, execution processes, metrics and the cross-functional alignment that a fast-growing organization needs. Your contributions will directly impact how fast our performance R&D gets productized. If you can drive turning a set of ambitious but loosely defined initiatives into a predictable, well-governed program, this role is for you.
EXAMPLE INITIATIVES
Take a look at these blog posts written by members of our Model Performance team:
- How to build a day-0 API for Kimi K3
- How we built the new fastest API for GLM-5.2
- Inference engineering for DeepSeek V4 Pro 0813
RESPONSIBILITIES
- Own execution across Model Performance's active project portfolio, freeing the team's technical leads to focus on technical direction rather than tracking.
- Design and stand up the planning structures, operating cadences, and status reporting mechanisms that best fits the team’s DNA.
- Coordinate model release and optimization programs end to end, including day-zero launches, sequencing the work across performance engineering, infra, and release stakeholders.
- Drive cross-team alignment as scope expands from Model Performance Core into Model APIs and the inference production stack (BIS).
- Surface risks and dependencies early, and keep leadership informed with clear, honest status.
- Partner with engineering leads to design team structures and ownership boundaries as the org scales.
REQUIREMENTS
- Deep technical program management experience - you're already running programs of this scope at an organization of similar or greater complexity.
- Experience program-managing model performance or inference optimization work - you understand how engines like vLLM, TensorRT-LLM, SGLang, or NVIDIA Dynamo fit into a production serving stack and can engage credibly with the engineers building on them.