SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
LiveKit is building infrastructure for the voice-driven era of computing, powering voice AI applications for OpenAI, xAI, Salesforce, Coursera, Spotify, and thousands of others, facilitating billions of calls annually.
You will own LiveKit Inference, a gateway providing developers managed access to the best models for voice AI through a single integration. This is one of LiveKit's most important revenue-generating products and requires a dedicated owner to establish foundational strategy and execution.
Key responsibilities include:
• Own the complete LiveKit Inference roadmap, setting vision and sequenced plans in partnership with engineering, deciding what gets built and when.
• Ship a model layer that developers reach for by default—making accessing the best models through LiveKit easier and better than integrating providers directly.
• Keep pace with the rapidly evolving model landscape by deciding which models, providers, and modalities to add and when, ensuring LiveKit stays ahead.
• Grow value delivered and captured by consolidating model spend onto LiveKit in ways that raise customer value without raising costs.
• Develop deep customer and data insights to understand how teams use models in production, shaping priorities accordingly.
You bring 8+ years of product management experience at senior or staff level, ideally having owned developer-facing APIs, platforms, or AI/ML products. You possess technical depth in model serving, routing, latency, and quality tradeoffs, earning trust from strong engineers without necessarily writing code. You've owned products where cost, performance, and vendor tradeoffs were central, and you can keep products current as landscapes shift. You influence through clarity, trust, and good judgment rather than process. You communicate exceptionally—writing clearly, presenting well, and moving fluidly between board-level strategy and detailed technical discussions. You're comfortable with ambiguity, figuring out what matters, prioritizing ruthlessly, and moving quickly without perfect information.
Nice-to-have experience includes building model gateways, inference APIs, model-provider products, or AI/ML platforms; familiarity with model benchmarking and performance metrics like TTFT; knowledge of STT, TTS, LLM, and realtime model spaces; background at developer tools, API infrastructure, or AI/ML companies; experience with voice AI, real-time communication, or telephony; and experience with usage-based pricing and cost/margin optimization.
