Decagon

Introducing Decagon Labs. The research team behind the models and infrastructure powering 80% of Decagon’s model traffic. More in the thread. ↓

The Enterprise Tech 30 list is out for 2026, and we’re proud to see four BCV portfolio companies recognized. From AI models to vertical SaaS, this year’s list spans the full enterprise stack across seven categories — a snapshot of how fast the landscape is evolving. @crosbylegal: The AI law firm built for execution with lawyer-assisted AI to review MSAs, DPAs and NDAs in under an hour. @DecagonAI: Building, optimizing, and scaling AI agents that treat every customer like the only one. @WeAreLegora: Building the world’s first truly collaborative AI for legal professionals. @mintlify: Helping teams create and maintain world-class documentation built for both humans and AI. Congrats to these teams pushing the boundaries of enterprise software. @Wing_VC

So excited to see 4 Chemistry companies on the Enterprise Tech 30 VC list! Huge congrats to @getserval, @meetgranola, @assort_health and @DecagonAI for their well-deserved recognition. Thanks to Wing and @EricNewcomer for the writeup! newcomer.co/p/mintlify-ser…

Many software products are adding AI copilots, mostly assistants that can find answers, take actions, etc. But just like how coding is moving to autonomous agents, a similar shift will happen in-product. We've pioneered the use of long-running agents via Decagon Duet. [1/4]👇

Introducing Decagon Labs, our research and agent orchestration team. Over 80% of model traffic at Decagon now runs on models we’ve trained in-house, structured as a network of specialized models handling different parts of the interaction: detection, orchestration, response generation, and evaluation.

Join our research team: decagon.ai/careers

More on our research: decagon.ai/blog/introduci…

Decagon Labs is focused on pushing that forward, from post training techniques to evaluation systems to new model architectures. The models coming out of Decagon Labs already power millions of customer interactions, delivering concierge level support across every conversation.

Instead of relying on a single model, we’ve built a network of specialized models, each responsible for a specific part of the interaction: detection, orchestration, response generation, evaluation. That separation lets us optimize each layer independently and drive better speed and quality across the system.

General foundation models are powerful, but they’re optimized for broad reasoning. Enterprise CX has a very different set of constraints: low latency, high accuracy, and consistent behavior across millions of interactions. Which is why over 80% of model traffic at Decagon now runs on models we’ve trained in house, purpose-built for enterprise CX.

Introducing Decagon Labs. The research team behind the models and infrastructure powering 80% of Decagon’s model traffic. More in the thread. ↓

@arianevans @monarch_money Happy to help! 🫡

@monarch_money Good news: your AI chat helped me figure it all out! (once past canned answers) Excellent AI chat help - possibly best I've encountered. Are you using an entry level frontier model, or did you fine tune and RAG an open model? Kudos to whomever did that work!

@monarch_money is there a way to review recurring merchants without getting blocked by requirement to bind them to an existing account (setup in Monarch)? Is there a way to setup recurring charges without binding them to an institutional account as key value?

Duet is now available to a limited group of customers. If you're interested in being among the first to build with it, reach out in the comments or drop me a message. decagon.ai/blog/introduci…

Duet distills what we've learned about conversational design, agent building, and hillclimbing from hundreds of enterprise deployments, and uses that expertise to compress days of manual work into minutes.

Introducing Decagon Duet, your AI partner to build self-improving AI agents. More in the thread. ↓

@a16z 🤝

In the 1990s and 2000s, email “replaced” physical mail and changed the nature of correspondence. The same thing is happening right now with customer attention. If you think AI is just going to “replace” the call center agent, you’re stuck thinking about first-order effects. AI is going to create an entirely new category of customer relationship that was previously impossible, just like how email created an entirely new category of communication. a16z's Sarah Wang on how Decagon is using AI to give everyone concierge service: a16z.news/p/the-internet…