---
title: "CypressConf 2026: Answering the signal"
url: "https://www.cypress.io/blog/cypressconf-2026-answering-the-signal"
description: "CypressConf 2026 is a wrap! Thank you to everyone who joined us for three days of talks and hands-on workshops. This year’s theme, The Signal Is Up, asked a…"
---

# CypressConf 2026: Answering the signal

October 26, 2026

By Melissa Milligan

CypressConf 2026 is a wrap! Thank you to everyone who joined us for three days of talks and hands-on workshops. This year’s theme, The Signal Is Up, asked a question every team knows well: how do you know a release is ready when the build is green? 

### **From the founders: Quality in the era of the agentic SDLC**

In the founder session, Brian Mann, Founder of Cypress.io, and Jennifer Shehane, Co-founder and Head of Product, talked about using AI so teams spend more time building and less time cleaning up. Jennifer showed how recent releases like Cypress Tap, the Cloud CLI, Cloud MCP and Webhooks give agents the context to test and debug on their own.

Brian looked at the road ahead. He estimated that most teams have handed only 5% to 20% of their work to agents so far, and that work needing judgment will be the hardest to hand off. For agents to do that work well they need access to your systems, an understanding of what your data means, and a way to tell whether they got it right. His closing point framed the whole conference. The job of software quality has always been deriving and validating intent: did we ship the thing we meant to ship? Agents change the form factor. They don't change the job. Brian has spent the past year exploring how teams can confirm that what they ship matches what they meant to build. Want to help shape it? [Join the design partner program](https://on.cypress.io/cyconf-dpp?ref=cypress-io.ghost.io).

### **Lessons from industry leaders**

Across the sessions, speakers kept coming back to the same idea: a signal is only useful if you can trust it. Here’s what we learned.

**Green is not the goal**

A passing suite doesn’t always mean a release is safe, because coverage numbers and pass rates can hide shallow assertions and flaky tests. Judge your tests by whether they would catch a real bug and whether you’d trust them to make the release call, and keep checking critical paths after they ship.

**Governing is the new writing**

Agents can write tests in seconds, so more of the work now goes into reviewing what they produce. Give your agents clear written rules, keep each task small enough to review properly, and keep critical tests in version-controlled code that a person has checked.

**Debugging at the speed of the signal**

Most of the time spent on a failed run goes into working out whether it’s a real bug, a known flake or an environment problem. Give people and agents the full picture, including test history, Test Replay and error patterns, and rerun only the tests that failed.

### **Catch the sessions**

All sessions will be published to the [Cypress YouTube channel](https://www.youtube.com/playlist?list=PLPPb-bn9jZmI&utm_source=blog&utm_medium=cypress.io&utm_content=cypressconf2026) in the coming weeks. In the meantime, put the Cypress AI features to work on your own runs and strengthen your signal. 

Thank you to our speakers and to everyone who kept the chat busy for three days. We can’t wait to see you next year!