Delivering high-quality, real-time audio and video is a growing challenge for tech companies. Especially as the demand for AI-native, multimodal experiences skyrockets. While some teams try to build their own streaming infrastructure. The reality is that maintaining low-latency performance across global networks is complex and resource-heavy. Enter LiveKit, the open-source communications platform quietly powering everything from 911 calls to ChatGPT’s voice mode.
Founded in 2021 by Russ d’Sa and David Zhao, LiveKit started as a tool to simplify building real-time voice and video apps. The idea quickly caught on. Today, the San Jose-based startup serves more than 500 paying customers and supports over 100,000 developers across its open-source and cloud offerings.
LiveKit has become a backbone of mission-critical systems. According to d’Sa, its technology now supports about 25% of U.S. emergency 911 calls. It’s also used by aerospace firms during rocket launches. Skydio for police drone operations, and by teams at Oracle and Adobe working on government-grade applications.
The turning point came when big tech companies like Spotify and Reddit began using the open-source version and asked for a managed cloud solution. That prompted d’Sa, formerly an early Twitter engineer, and Zhao, a former Motorola engineering director. To officially turn LiveKit into a commercial venture with the launch of LiveKit Cloud.
In simple terms, LiveKit is like Cloudflare—but built for media streaming. Its SDKs, APIs, and infrastructure tools let companies build robust, real-time audio and video experiences at scale.
Some of the world’s biggest tech players are now LiveKit customers—including Meta, Microsoft, and Spotify. As well as emerging AI startups like Character AI and Speak. OpenAI even uses LiveKit to power the voice capabilities in ChatGPT.
As the market shifts toward AI-driven interactions, LiveKit is positioning itself for the future. Its newest initiative is an “elastic agent compute service” designed to dynamically scale real-time AI voice agents—think chatbots that talk—up or down as needed.
“What Stripe did for payments, we’re doing for communications,” d’Sa explained. “LiveKit is becoming an AI-native cloud platform.”
That vision has investors lining up. LiveKit recently secured $45 million in Series B funding, led by Altimeter Capital. Redpoint Ventures and Hanabi Capital also joined the round. With this new backing, the company plans to grow its 50-person team, scale its infrastructure, and double down on product development.
Financially, the startup is already on solid ground. LiveKit crossed a $10 million run rate last year and continues to grow quickly, fueled by demand from developers building next-gen, multimodal applications.
In a world where real-time communication is increasingly key to AI interfaces, LiveKit is quietly becoming the platform behind the voices—and videos—of tomorrow.
