AI concierge agents for enterprise customer experience — build, optimize, and scale support that treats every customer like the only one
HQ: San Francisco, CA, United States | Founded: 2023 | Employees: ~200 (2026, est.) | Stage: Late Stage VC (Series D) | Website: https://decagon.ai
Decagon is an enterprise AI company that builds autonomous AI agents for customer experience, founded in August 2023 by Jesse Zhang (CEO) and Ashwin Sreenivas (CTO), two engineers who previously led AI and machine-learning work at Uber. The company emerged from stealth in June 2024 with $35M in funding from Accel and Andreessen Horowitz, positioning itself against the incumbent chatbot market it considered fundamentally broken.
The platform lets brands deploy AI agents across voice, chat, email, and SMS that resolve issues, take actions, and complete tasks on behalf of customers 24/7. Its differentiator is the Agent Operating Procedures (AOPs) framework: CX teams write agent behavior in natural language — like how humans learn standard operating procedures — and those instructions automatically compile into code that production systems execute, giving business users hands-on control while technical teams retain full control over guardrails and integrations.
Traction has been exceptional. Customers including Duolingo, Notion, Block, Avis Budget Group, Deutsche Telekom, Oura, and ClassPass report average deflection rates near 70% (Duolingo above 80%), a 95% reduction in the cost of support conversations for ClassPass, and a 3x CSAT improvement for Oura. The company grew from zero to 8-figure ARR in its first year and crossed $100M in annualized revenue in 2026, up roughly 3x from ~$44M at the end of 2025. Total funding stands at $481M, capped by a $250M Series D at a $4.5B valuation in January 2026 led by Coatue and Index Ventures — making Decagon one of the fastest enterprise AI valuations in history.
Decagon is the fastest-scaling AI agent company in customer experience: founded in August 2023, it reached $100M in annualized revenue by mid-2026 and a $4.5B valuation in January 2026. Its edge is a product-driven approach — Agent Operating Procedures (AOPs) that compile natural-language instructions into code — which lets CX teams iterate on agents without professional-services armies, and its willingness to bet on outcome- and volume-based pricing. The bull case is that agentic CX is one of the largest near-term enterprise AI markets and Decagon has marquee logos (Duolingo, Notion, Block, Avis, Deutsche Telekom) plus 70%+ average deflection rates; the bear case is intense competition from Sierra, Intercom, Zendesk, and Salesforce Agentforce, and dependence on frontier LLM providers for core intelligence.
Browse: Home | Blog | About | All Startups A-Z | View full profile on StartupWiki