Dennis Cui

VP Eng @DecagonAI

[Dennis Cui @dennis_cui]

So excited to announce Alto, our newest voice model. Alto is fast, natural and smart. It responds in under 700ms and can guide users through the most complex workflows. Customers can no longer distinguish Alto from human agents.

Shoutout to the team at Decagon Labs.

[Decagon @DecagonAI]

Introducing Alto, our new voice model from Decagon Labs.

Alto powers our duplex architecture which lets agents listen, speak, and act at once. It delivers a 5+ percentage point lift in resolution with ~700ms responses on live calls.

[Dennis Cui @dennis_cui]

This is how personal agents will get work done reliably

[Noah Shinn @noahrshinn]

Native customer service in Instinct

Nobody likes talking to customer service. Instinct should handle those tasks for you.

With Instinct Concierge, we enabled Instinct to make calls on your behalf. Now, we’re taking this a step further. When your Instinct connects with participating businesses on @SierraPlatform and @DecagonAI, the agents will be able to work together to solve your problem with speed, reliability, and safety without making a phone call.

I’m excited to announce that Instinct will be partnering with Sierra and Decagon to bring customer service experiences directly to your personal agent.

Customer service tasks like:
- Troubleshooting your home wifi issues
- Finding out when a delayed package will arrive
- Figuring out why your bill is higher than last month
- Getting clarity on a confusing return or cancellation policy

[Dennis Cui @dennis_cui]

Excited to partner with @instinct and @Muse to safeguard consumer <> business touchpoints

[Decagon @DecagonAI]

Today, we're open-sourcing the Personal Agent Consent & Trust Protocol (PACT), co-developed by @Instinct. We're also joining the Personal Agent Protocol working group by @Muse.

Built on A2A and OAuth, PACT enables businesses to verify who a personal agent is acting for and what the customer has authorized it to do.

Excited to shape the industry standard for how personal agents will interact with businesses.

[Dennis Cui @dennis_cui]

We know AI tools can significantly boost productivity. But, engineers can multiply those gains by debugging their own workflows and removing bottlenecks:

What about the codebase made it difficult for the agent to solve a task?
What proof could the agent provide to make verification easier?
What value did human review add? Why didn’t your agent provide the same feedback?

Use the answers to improve code quality, docs, tests, and review process.

As the tools improve, what’s slowing you down will change. Strong engineering teams keep looking for the next bottleneck and removing it.

[Dennis Cui @dennis_cui]

One of my favorite interview questions is: "Teach me about something you've worked on."

What I'm looking for is how deeply someone has tried to understand the things they've worked on. You can usually tell pretty quickly who is genuinely curious.

[Dennis Cui @dennis_cui]

The team has been experimenting with Devin recently, with promising results. Excited to see where @Cognition takes it. There's still so much room to build what developers actually need to get more done.

[Dennis Cui @dennis_cui]

In July, 80% of our team's AI coding happened in Claude. By August, 80% was happening in Cursor or Codex.

One of the most valuable traits for engineering teams right now is a willingness to abandon a workflow they relied on just weeks ago.

[Dennis Cui @dennis_cui]

@spinnercat @yak123soba oh that

[Dennis Cui @dennis_cui]

@yak123soba @spinnercat what's beautiful?

[Dennis Cui @dennis_cui]

@MatthewBerryTMR @JosinaAnderson Frank Gore or Dan Herron?

[Dennis Cui @dennis_cui]

@spinnercat @MaryMeerkatZhu @markyao6275 don't you only need 5...

[Dennis Cui @dennis_cui]

@MaryMeerkatZhu @markyao6275 @spinnercat or you could switch to CS no problem. No more pages of hexagons.