The Tester’s Guide to Dreamforce 2025
Okay, quick truth: almost every Dreamforce post forgets QA. They gush about demos, agent magic, and feature drops; then toss a line about testing like it’s optional. We don’t do optional. We make stuff work. So we rewrote the narrative: at Dreamforce 2025, QA isn’t the quiet cousin; QA is the headline act. This is your conversational, no-fluff playbook for owning the conference ; from keynotes to the campground ; and yes, for swinging by TestZeus at Booth #351.
Why we care (and why you should too)
Salesforce is rolling out Agentforce, AI-first flows, and real-time integrations; exciting stuff, but also a fresh pile of failure modes. Every demo that gets applause is backed by a test someone wrote, or it should be. Our job? Turn hype into repeatable, trustworthy behavior. Dreamforce is where the roadmap drops; we’re there to translate the roadmap into test plans and deployable QA strategy.
The vibe: sniff for risk, not soundbites
When you’re listening to a keynote or watching a demo, don’t treat it like theatre ; treat it like reconnaissance. Ask yourself and the speakers small, sharp questions:
What could make this fail in production?
How would I prove it works for our customers?
What observability will be available when things go sideways?
If the demo can’t show logs, traces, test harnesses, or rollback hooks ; you just found a gap.
The hands-on beats the hype (Trailblazer Forest checklist)
Walk the booths and demo lanes like a detective:
Developer Grove: watch for AI-assisted code paths and ask how generated code is validated. Bring a keyboard of skepticism.
Admin Meadow: watch deployment/automation demos; ask about sandbox parity and data masking.
Camp Mini Hack: join a challenge; it’s fast insight into developer assumptions and edge cases.
Quick playbook: work the conference like a QA lead
Bring two real problems from your org. Real examples get concrete answers.
Ask for artifacts: logs, model versions, replay tools, and test harnesses. If someone hands you a demo without reproducible evidence, push harder.
Probe governance: who chooses the model; how are bias and drift monitored; what’s the rollback plan?
Validate sandboxes: can you run safe tests in various sandboxes? can you clone production scenarios without exposing PII?
Collect measurable KPIs: precision, false-positive rate, drift rate, latency budgets ; don’t leave these fuzzy.
Sessions worth your time (and the one-line reason why)
Agentforce / Developer Keynote - sees the new attack surface; valuable for test scoping.
Testing Center / Scale Center demos - where load and agent testing may finally get a toolkit.
True to the Core - ask the hard question about sandboxes and testability.
Platform Park - watch integration demos and note stateful scenarios you’ll need to automate.
Here are the ones we have added to our agenda:
Networking, but smarter
Stop collecting stickers and start collecting answers. When talking to vendors or PMs:
Show your two test stories up front.
Ask them to walk you through a failure scenario and how they'd detect it.
If they offer a product, ask for a short POC focused on your failure mode, not a generic demo.
Pro tip: swap contact info by promising a follow-up with a failing test case; it’s the fastest path to concrete commitments.
Stop by TestZeus Booth #351
(Main Campground) ; Don’t show up empty-handed
We’ll be at Booth #351 in the main campground; here’s how to make that visit count:
Bring a 2-minute test story (a flaky Flow, a broken integration, a model that misfires). We’ll show you how our agent-led test workflows would catch it.
Ask for a live demo showing how TestZeus creates and runs a targeted end-to-end test; get them to show logs and the failing step replay.
Scan the QR on our booth for an immediate take-home report template you can run after returning to your org.
Book a short slot with our engineers if you want to deep-dive; Booth #351 will have limited guided demos ; reserve early.
Swag & follow-ups: pick up a TestZeus sticker, then ping us on Slack/X with your test story so we can follow up with a tailored pilot plan.
We’re bringing a few hands-on demos focused on agent testing, sandbox cloning for safe load, and a sample Agentforce Testing Center run ; so yes, come ready.
Leave with outcomes, not just swags
Make Dreamforce worth it: return with a prioritized list of gaps, three vendor/PM follow-ups, and one tactical pilot you can run inside 30 days. Show your team the test stories you collected and the specific fixes or pilot projects you negotiated at the show.
If the wider conference forgets QA, we won’t. We’ll be the ones people call when a demo breaks; the ones who can reproduce, trace, and mitigate issues. Show up curious, guarded, and ready to flip the script.
See you at Booth #351 ; come say hi, tell us your worst bug, and let’s fix it together.
// Start testing //
