Technical Program Manager, Quality

Sesame•San Francisco, CA

About The Position

At Sesame, we are building the next generation of voice-based personal agents. We are seeking a Technical Program Manager to bridge the gap between LLM research and a polished, user-ready product. In this role, you will drive the development lifecycle, ensuring novel capabilities are delivered on schedule and rigorously measured against real-world performance. You will also be responsible for the "quality loop": transforming qualitative user feedback into technical evaluation datasets and ensuring our models evolve based on measurable data.

Requirements

Academic Background: BS or MS in Computer Science, Electrical Engineering, or a related technical discipline.
TPM Experience: 3+ years of experience managing LLM research or AI/ML product cycles.
LLM Foundations: A solid technical understanding of LLMs (training, prompting, inference, quality) and the unique constraints of voice-based interfaces.
Programming Skills: Proficiency in Python with the ability to independently write and implement evaluation logic and data-handling scripts.
Communication: Proven ability to coordinate and align with stakeholders across different functions, managing complex dependencies with great technical depth.

Nice To Haves

AI Voice Experience: Prior experience with LLM-based voice products and modern generative speech technologies.
Automated Testing: Experience integrating model evaluation workflows into automated testing and deployment pipelines.
System Logic: Familiarity with the architectural challenges of deploying low-latency, conversational AI agents.

Responsibilities

Project Orchestration: Manage the end-to-end lifecycle of LLM projects, navigating the transition from research milestones to production-level deployments.
Metric Definition & Design: Transform subjective user feedback into objective metrics and datasets.
Eval Implementation: Design and implement technical evaluations to address issues found in the field. You will be expected to dive into the codebase to help integrate these evals into existing pipelines.
Feedback Loop Management: Track internal and external feedback, ensuring identified issues are followed through to resolution in subsequent iterations.
Release Management: Maintain the technical roadmap for voice-based capabilities, proactively identifying dependencies and resolving technical blockers across teams.
Cross-Functional Alignment: You will ensure the roadmap incorporates the specialized work and constraints of all teams—not just ML—to deliver a cohesive user experience.