Meet TestDino at Dreamforce 2026
AI agents write code faster than your team can verify it. TestDino records every Playwright run from your CI, so every pull request, human or agent, clears the same check before merge.
Meet us off the expo floor
Dreamforce fills three city blocks for three days. Book a working session near Moscone instead of a badge scan.
On the ground September 15-17.
45 to 60 minutes, near Moscone.
Your CI pipeline and Playwright coverage.
Written plan and a trial workspace to keep.
attendees at Dreamforce
sessions in the catalog
on the ground
private working sessions
Where Salesforce testing is heading
Six shifts hitting Salesforce QA teams at once. Bring one to a whiteboard session.
Agent evals
Agentforce Testing Center scores utterances in a sandbox. Your agent's behavior in the live app still needs end-to-end checks.
Judges need ground truth
Judge scores drift without deterministic pass/fail signals underneath. Playwright assertions give you that baseline.
Shadow DOM automation
Selenium-era locators fight Lightning Web Components. Playwright pierces shadow roots natively.
Migration deadline: Dec 31, 2026
A forced migration is a free re-evaluation. Teams are moving to open-source Playwright with a reporting layer on top.
The storefront is a web app
Composable Storefront on PWA Kit runs on SCAPI and OCAPI. Playwright covers the UI and the APIs in one run.
Three releases a year
Winter, Spring, and Summer releases force regression on Salesforce's calendar, not yours. Data Cloud changes ride along.
One gate between code and merge
TestDino sits between your CI and your merge button. Each capability links to how it works.
Three days on the ground
Sessions run 45 to 60 minutes near Moscone. Each day has a default theme, not a fixed one.
Tuesday
Agents and evals
Map your Agentforce test coverage and design the eval layer under it.
Wednesday
Test architecture
Whiteboard your CI pipeline, LWC automation, and a Provar exit plan.
Thursday
Commerce, data, and releases
Storefront coverage, Data Cloud validation, and a three-release regression plan.
What we take in the session
Sessions run 45 to 60 minutes near Moscone. Bring the one that is stuck.
Each release still means days of manual clicking.
Your renewal is on the table before Dec 31, 2026.
A pilot is live with no eval plan behind it.
Coding agents ship faster than you can verify the runs.
Leave with a 30-day plan
Every working session ends with a written 30-day plan and a free trial workspace. Both are yours to keep.
- 1
Week 1
Connect CI and baseline your pass rate and flake map.
- 2
Week 2
Set up trace-based triage and quarantine known flakes.
- 3
Week 3
Wire the MCP server into Claude Code or Cursor.
- 4
Week 4
Review trends and set your release gate.