Open source infrastructure for voice, video and physical AI agents, the stack behind ChatGPT's voice mode.
LiveKit has raised about $181M. Redpoint led a $7M seed in December 2021 with Justin Kan, Robin Chan and Elad Gil; Altimeter led a $22.5M Series A in June 2024 with Redpoint and a long angel list including Jeff Dean, Aravind Srinivas, Amjad Masad, Guillermo Rauch, Mati Staniszewski and Demi Guo, at a reported $110M valuation; Altimeter led again on a $45M Series B in April 2025, joined by Redpoint and Mike Volpi's Hanabi Capital in its first investment, at a reported $345M valuation and $83M total raised. On January 22, 2026 the company announced a $100M Series C led by Index Ventures (Sahir Azam joined the board) with Salesforce Ventures, Hanabi, Altimeter and Redpoint, at a $1 billion valuation. Run rate passed $10M in 2024 with over 500 paying customers by April 2025, and in August 2023 the founders turned down an acquisition offer of roughly $20M, the same day OpenAI first emailed about working together.
LiveKit is the infrastructure layer for real-time voice, video and physical AI agents. It started in July 2021 as an open source WebRTC media server written in Go (Apache 2.0, about 21,000 GitHub stars) and grew into a full platform: the LiveKit Agents framework for building voice agents (Python, about 14,000 stars, over a million downloads a month), LiveKit Cloud for running them, plus Inference, Telephony with native Phone Numbers, Agent Builder, serverless agent hosting and agent observability. A developer can switch between self-hosting and Cloud without changing a line of code, which was the one design rule the founders set when Cloud launched in 2022.
More than 300,000 developers and 5,000 companies run production workloads on it, collectively handling billions of calls a year. OpenAI uses LiveKit for ChatGPT's voice mode, and other named customers include xAI, Salesforce's Agentforce, Tesla, Spotify, Meta, Coursera and Character.ai; roughly a quarter of US 911 dispatch centers also run on it. Growth has been almost entirely bottom-up: Index Ventures wrote that it happened "with barely any GTM machinery behind it", and the company says many of its largest customers started as a single developer trying the free tier.
Russ d'Sa, CEO and co-founder. Born in Canada and raised in the Bay Area, he went through Y Combinator's summer 2007 batch with MeetYou, then was an engineer at 23andMe (2008 to 2009) and Twitter (2009 to 2012), which he joined as its second front-end engineer. He was the first employee at Expa in 2013, co-founded Evie Labs the same year and stayed through its 2019 acquisition by Medium, where he ran consumer product until 2021. BS with honors and MS in Computer Science from UC Davis.
David Zhao, CTO and co-founder. Built internet-scale services at Amazon, then co-founded Zecter (ZumoDrive) in the same 2007 Y Combinator batch as d'Sa; Motorola Mobility acquired it in December 2010 and he became a director of engineering there. Co-founder and CEO of Evie Labs from 2013 until Medium acquired it in 2019, then director of engineering at Medium until 2021, and a senior venture partner at Pioneer Fund. BS in Computer Engineering from the University of Washington. The two have worked together for nearly two decades.
LiveKit is headquartered in San Francisco at 369 Sutter Street, an office set up in 2026 for its hybrid teams, and hires remotely across the United States, Canada, the United Kingdom, Ireland, the European Union, India and the wider APJ and EMEA regions, with hybrid roles in San Francisco and Austin and in-person forward deployed roles in New York and several other US cities. Most roles are remote within the United States, several engineering roles are open across North America, APJ and EMEA, forward deployed and developer success roles are listed for the EU and the UK with local pay bands, and a Spanish/English forward deployed role is tied to Miami, Austin, Chicago, Los Angeles or San Antonio. Its staff are spread across 14 countries, and funding announcements through 2025 described the company as San Jose-based, before the San Francisco office opened.
Describe your next role, cut the noise
Own brand campaigns, events, presentations, environmental design, and swag for an AI developer platform; must have senior brand design experience across digital, print, out-of-home, environmental, presentations, campaign production, systems thinking, AI tools; bonus: motion, developer-facing/design-forward experience; equity
Remote US only
Owns the GTM financial engine for a metered AI infrastructure business scaling several-fold in 12–18 months; partners directly with CRO and VP of Finance; owns consumption revenue forecasting, deal desk, pricing and margin governance, and sales compensation; builds the GTM finance model from pipeline to bookings to recognized consumption revenue, including ramp/utilization assumptions, annual planning, and KPI analytics; must have 6–10 years in GTM strategic finance/FP&A/RevOps/deal desk at high-growth B2B tech companies plus 2+ years in investment banking and/or consulting; must have expert Excel/Google Sheets modeling, deep GTM metrics/economics knowledge, and hands-on quota/comp plan/sales capacity modeling; bonus: voice AI/CPaaS/telecom/real-time infrastructure, usage-based or credit-denominated business models, early-stage startup experience, AI tools at work, CRM/billing/commission tooling, ERP familiarity
Remote US only
Build and own security across applications, services, infrastructure, and developer workflows for a platform used to build, deploy, scale, and observe AI agents in production; hands-on engineering role focused on secure systems, code, tooling, automation, threat modeling, secure code reviews, architecture discussions, auth/access control hardening, vulnerability investigation, incident response, and disclosure processes; requires 6+ years as a software engineer with strong security engineering focus and experience across application and/or infrastructure security, CI/CD, cloud platforms, containers, and real-world security incidents/postmortems/remediation; bonus: security reviews for WebRTC/media pipelines/real-time systems, open-source security tooling or research, static/dynamic analysis, fuzzing, sandboxing, prior LiveKit usage
Remote in North America, APJ, and EMEA
Own social and blog presence aimed at developers across X, Reddit, TikTok, LinkedIn, and blog; turn product launches into developer-focused social campaigns; participate directly in agentic AI conversations and tie engagement metrics to developer awareness/adoption; close collaboration with engineering, product, and DevRel; must have proven track record growing a developer audience on social, exceptional writing across platforms, strong interest in agentic AI, and data-driven experimentation; bonus: interest in voice AI or experience building with a voice agent, personal developer following, experience at an AI or developer-tools company; open source contribution mentioned
Remote US only
The role requires 7+ years of experience in backend or infrastructure engineering with proficiency in languages like Go, Python, or Rust. Candidates must be fluent in both Spanish and English and possess strong skills in distributed systems and real-time media.
Remote US only; listed in Miami, Austin, Chicago, Los Angeles, or San Antonio.
Requires 4+ years of end-to-end B2B SaaS demand generation or growth marketing experience. Candidates must be data-driven and proficient in marketing automation tools like HubSpot, Marketo, or Salesforce.
Remote in the United States only; no other geographic regions specified.
The role requires 4+ years of experience in IT engineering, systems administration, or identity engineering with hands-on experience in Okta and MDM solutions. Candidates must demonstrate strong ownership, excellent writing skills, and the ability to build and ship technical solutions quickly.
Remote North America only
The ideal candidate is a strong Python engineer with proven experience taking models from raw data through to production deployment. You should be comfortable working with GPUs, designing reward systems, and collaborating effectively in a remote environment.
Remote in North America, APJ, or EMEA
Candidates must have 8+ years of product management experience, specifically in developer-facing or usage-based products. A strong background in product-led growth, technical proficiency, and the ability to influence cross-functional teams is required.
Remote in the United States (or SF HQ).
The role requires 8+ years of product management experience, specifically with developer-facing products and deep technical understanding of telephony systems. You must demonstrate strong cross-functional influence, reliability instincts, and the ability to operate in an ambiguous, fast-paced environment.
Remote in the United States (or in-office at SF HQ).
Candidates must have experience in partnerships, alliances, or technical business development, ideally with GSIs or technology partners. The role requires the ability to operate with ambiguity and possess enough technical credibility to engage effectively with engineers and technical stakeholders.
Remote in the United States only; no other regions mentioned.
The ideal candidate has a proven track record of building enablement functions at high-growth SaaS companies and possesses the ability to translate deeply technical concepts for sales teams. You must be comfortable working in an ambiguous, fast-paced environment and be willing to travel monthly for in-person team activities.
Remote within the United States; requires monthly travel for in-person enablement.
Requires 8+ years of senior-level product management experience, specifically with developer-facing infrastructure or platform products. Must possess deep technical knowledge of distributed systems, scaling, and reliability to effectively partner with engineering teams.
Remote in the United States.
Requires extensive experience designing and delivering production-grade distributed systems with a strong grasp of consensus and failure modes. Proficiency in Go and experience with low-latency infrastructure is highly desired.
Remote in the United States only.
Requires a practicing engineer capable of building and debugging production software across unfamiliar codebases. Must possess excellent communication skills and a strong customer-centric mindset.
Remote Ireland (EU) only
Requires 5+ years of experience in backend or infrastructure engineering with proficiency in languages like Go, Python, or Rust. Candidates should be comfortable working with distributed systems and have strong communication skills for customer-facing technical leadership.
Remote from Ireland (EU)
Requires 8+ years of experience in technical customer-facing roles with fluency in developer-platform fundamentals and distributed systems. Candidates should have a proven track record of being a trusted technical advisor for engineering teams and executives.
Remote US only; occasional on-site visits for launches and executive reviews.
Requires over 5 years of experience in backend or infrastructure engineering with proficiency in languages like Go, Python, or Rust. Candidates should be comfortable working with distributed systems and have strong communication skills for customer-facing technical leadership.
Remote UK-based; on-site in London available as needed.
Requires a practicing engineer capable of debugging production software and communicating complex technical concepts to diverse audiences. Candidates should be comfortable navigating unfamiliar codebases and possess a strong customer-centric mindset.
Remote UK only
Requires experience as an early in-house lawyer at a high-growth tech company or a senior commercial counsel. Must be fluent in GDPR, CCPA/CPRA, and commercial contracts for enterprise and data-heavy products.
Remote US only
Requires deep, hands-on Salesforce expertise and solid HubSpot experience, treating CRMs as engineering systems rather than admin tools. Must have proven experience building AI agent workflows in code and managing complex API integrations.
Remote US only
Requires experience designing and delivering production-grade distributed systems with a strong grasp of consensus and failure modes. Proficiency in Go and experience with low-latency infrastructure or coordination systems is highly desired.
Remote US only
Requires 6+ years of engineering experience with strong proficiency in TypeScript and a track record of shipping user-facing software for technical users. Candidates must possess strong product sense and a genuine interest in agentic AI and developer platforms.
Remote US only
LiveKit is the plumbing under the most visible voice product in the world, and it got there sideways. Russ d'Sa built a voice interface demo for ChatGPT before OpenAI had shipped one; four months later OpenAI found it, having quietly built its own voice-mode prototype on LiveKit from a personal Gmail account, and reached out. That partnership grew into Advanced Voice Mode and vision, and now the same stack sits under Tesla's sales and roadside-assistance agents and Salesforce's Agentforce voice agents. The company's stated goal is to "make building and scaling voice AI as easy as building and scaling on the web".
The team is still small for that footprint. In April 2025 it was about 45 people, 37 of them engineers, and LinkedIn now lists 100 to 150 employees across 14 countries after the Series C. Postings are blunt about what that means: "At LiveKit, the infrastructure is the product, you're not building the layer underneath, you are the layer," and "The team is small. Your decisions ship directly into production." Many open roles are the first of their kind: the first product manager for telephony, the first technical account manager, the first sales enablement lead, the first data hire.
Every LiveKit posting, from Head of Legal to Staff Engineer, carries the same five-line test: you "obsess with crafting code that is fast, reliable and practical for the problem", are "known as the go-to person for tackling tough technical problems", "work hard and can build and ship fast", "can clearly explain complex technical concepts to others" and are "a fast learner, frequently picking up new languages and tools". Engineering postings then add: "The best way to impress us is with thoughtful Issues and/or PRs on our Github repos." Non-engineering postings say the same thing in their own words: tinker with the product before you apply.
Open source is the founding conviction rather than a marketing layer. David Zhao wrote at launch in 2021: "From ZumoDrive to Evie, I've benefited tremendously from open source software my entire career. I'm glad that I have a chance to give back now with LiveKit." When Cloud launched, the design imperative was that "no developer would be locked into our platform", and d'Sa's Series A line was "We'll continue working hard to build muck, so you don't have to."
The company describes itself as AI-first in how it builds. A 2026 data posting says the team works "in an AI-first development environment where tooling like Claude Code and skills-based workflows are core to how we operate", expects hires to use AI coding assistants extensively and to write skills and documentation "that work for both humans and AI agents", and the brand designer posting asks for someone "actively experimenting with AI tools as part of your creative process".
LiveKit applications go through Ashby, and the loop for engineers is long and practical. A software engineer who interviewed in early 2026 reported a recruiter screen, a technical interview and a take-home over about five weeks, and called it "more rigorous than anything I'd done at other big tech companies": the take-home was to build a production voice agent with sub-500ms latency, and the recruiter's preparation advice was, unusually, accurate. Infrastructure postings list the stack they will probe (Go, gRPC, Raft, NATS, Kubernetes, consensus and coordination) and note that Go fluency is a nice-to-have rather than a requirement, while product engineering roles are TypeScript-first with Go as a plus.
Experience bars are stated per role: 4+ years for IT and demand generation, 5+ for forward deployed engineers, 6+ for product, security and GTM finance roles, 8+ for technical account managers and staff product managers. Accommodation requests go to recruiting@livekit.io.
As of September 2026 LiveKit's postings list, in dollars for US roles and in euros and pounds for EU and UK roles: Staff Software Engineer, Distributed Systems $200,000 to $300,000; Distributed Systems Engineer, Product Engineer, GTM Systems Engineer and Head of Legal $120,000 to $250,000; Staff Security Engineer and Research Engineer $135,000 to $300,000; Forward Deployed Engineer $180,000 to $320,000 in the US, €153,000 to €274,000 in the EU and £126,000 to £185,000 in the UK; Developer Success Engineer €168,000 to €213,000 in the EU and £145,000 to £185,000 in the UK; Staff Product Manager $225,000 to $265,000; Technical Account Manager $170,000 to $200,000; Senior Brand Designer $145,000 to $187,000; IT Engineer $155,000 to $180,000; GTM Strategic Finance Manager $190,000 to $250,000; Partnerships Manager $336,000 to $383,000.
The core server and cloud control plane are Go with psrpc, gRPC, Raft, NATS, Kubernetes, CockroachDB, Kafka and Nebula, observed with Prometheus, OpenTelemetry and ClickHouse; the Agents framework is Python, client SDKs include Rust, and product surfaces such as the dashboard and CLI are TypeScript. Cloud runs across a blend of providers rather than one. Internally the company runs on Okta, Workspace ONE, Google Workspace and Slack on a macOS-first fleet, with Salesforce as the GTM system of record and HubSpot for marketing, and internal systems are in audit scope for SOC 2, ISO 27001 and PCI.
LiveKit sells LiveKit Cloud, a metered service billed across agent hosting, inference, telephony, media transport and observability, on top of the free Apache 2.0 open source stack that anyone can self-host. There is a free tier, and a Startup Program waives the $500-a-month platform fee for a year and adds $10,000 in cloud credits and $7,000 in inference credits spread across speech-to-text, text-to-speech and LLM providers.
A 2026 software engineering candidate reported: tell me about yourself, why LiveKit, tell me about a time you made a mistake, how would you design a support agent, and design a support agent that can help customers and process refunds. Non-engineering roles are told the best way to impress is a point of view on a real LiveKit problem, such as how you would ramp account executives on the product or approach one of its partners.
No. There are no internship or junior listings, and every engineering posting asks for at least five or six years of experience.
Want remote roles like these in your inbox? Tell us what you are looking for.
Describe your next role, cut the noise