Staff Product Manager, Inference

ABOUT LIVEKIT LiveKit is building the infrastructure layer for the voice-driven era of computing. Our platform gives developers everything they need to build, test, deploy, scale, and observe agents in production. Founded in 2021, LiveKit powers voice AI applications for OpenAI, xAI, Salesforce, Coursera, Spotify, and thousands of others, collectively facilitating billions of calls each year. ABOUT THIS ROLE We're looking for a Product Manager to own LiveKit Inference: the gateway that gives developers managed access to the best models for voice AI, through a single integration. Our PM team is small and high-leverage, with each PM coming in to establish the foundations in a new area of the business. Inference is one of the most important revenue-generating products we have. It needs a dedicated owner, and that would be you. You'll set the vision and roadmap for LiveKit Inference, from which models and providers we offer to how we think about the space evolving strategically over the medium and long term. You’ll partner directly with the engineering team that builds it. This product involves both commercial relationships with frontier model labs as well as direct model hosting and optimization: we recently launched a voice-optimized deployment of Gemma 4 31B, and we're scaling up to become the go-to inference provider for voice workloads. The goal is to make LiveKit Inference the obvious default over wiring up model providers directly, easier, faster, cheaper, or better, and eventually good enough that companies come to LiveKit for voice inference itself. The model landscape moves faster than almost anything in software, and you'll decide how LiveKit stays ahead of it. This role can be based remotely or in our SF HQ, where you'll be part of a collaborative in-office culture! YOU'LL THRIVE HERE IF YOU: - are obsessed with making the best product for our customers - are known as the go-to person for tackling tough problems - work hard and can build and ship fast - are focused on polish, detail, and quality - are a fast learner, frequently picking up new tips, tricks, and skills The best way to impress us is by trying LiveKit's product, understanding the space, and bringing smart thoughts and opinions to the table. WHAT YOU'LL DO - Own the LiveKit Inference roadmap. Set a clear vision and a sequenced plan for the product, and decide what gets built and when, in partnership with engineering. - Ship a model layer which developers reach for by default. Make accessing the best models through LiveKit easier and better than integrating providers directly, and keep raising the bar. - Keep pace with the model landscape. Decide which models, providers, and modalities we add and when, so LiveKit stays ahead of a fast-moving space. - Grow the value we deliver and capture. Consolidate model spend onto LiveKit in a way that raises the value customers get without raising their cost. - Get deep with customers and data. Understand how teams actually use models in production, and let that shape priorities. WHO YOU ARE - 8+ years in product management, at a senior or staff level, ideally having owned a developer-facing API, platform, or AI/ML product. - Technical depth. You can reason about model serving, routing, latency, and quality tradeoffs, and earn the trust of strong engineers. You don't need to write code, but you need to understand systems well enough to make good tradeoffs. - Infrastructure and tradeoff sense. You've owned a product where cost, performance, and vendor or provider tradeoffs were central, and you can keep a product current as the landscape shifts. - Cross-functional influence. LiveKit's engineering teams operate with high autonomy. You influence through clarity, trust, and good judgment, not process or hierarchy. - Communication as a superpower. You write clearly, present well, and can move between a board-level strategy conversation and a detailed technical discussion with an engineering lead. - Comfort with ambiguity. We're a small product team at a fast-growing company. You'll figure out what matters, prioritize ruthlessly, and move quickly without perfect information. NICE TO HAVE - Built a model gateway, inference API, model-provider product, or AI/ML platform - Familiarity with model benchmarking and how both performance and quality are measured (e.g. TTFT) - Familiarity with the STT, TTS, LLM, and realtime model space and top providers - Background at a developer tools, API infrastructure, or AI/ML company - Familiarity with voice AI, real-time communication, or telephony - Experience with usage-based pricing models and cost or margin optimization OUR COMMITMENT TO YOU - An opportunity to build something truly impactful to the world - Contribute to open source alongside world-class engineers - Competitive salary and equity package - Health, dental, and vision benefits - Flexible vacation policy LiveKit is an equal opportunity employer and does not discriminate on the basis of any characteristic protected by applicable law. If you require a reasonable accommodation during the application or interview process, please contact recruiting@livekit.io.