
hey everyone 👋 we're Sneha and Anushka, co-founders of Spur.
Spur is building the validation infrastructure for software teams. We started with web testing, and today we're launching native app testing across iOS and Android.
Spur's agents test your apps the way real users do. You define the behavior you want to test in natural language, and Spur figures out how to execute and validate it across iOS and Android, native interactions, and web views.
Teams at HelloFresh, Nextdoor, DocuSign, Bilt, Alo Yoga, and Abercrombie are already using Spur for testing across their digital experience – on websites and on their apps.
Our Ask: If you’re building and maintaining a mobile app, we'd love to help you test it.
Book a demo: https://spurtest.com/book-a-demo
Software is getting generated faster than ever: features that used to take days to build are now generated in minutes. But - once the feature exists, someone still has to figure out whether it actually works.
That validation process hasn't sped up at the same rate.
If you’re shipping mobile apps, this gets even harder. The same product flow often turns into separate automation for ios and android. Teams maintain the tests, the device infrastructure they run on, and the CI setup around them.
And as your product changes, your automation has to change with it.
Natural Language allows you to define the “intent” of a test:
Most test automation describes exactly how to use an application.
Find this element. Tap it. Wait for another element. Enter text. Assert a value.
That makes the test tightly coupled to the implementation. Spur tests describe what the user should be able to do instead.
Add an item to cart, apply a promo code, and complete checkout with Apple Pay.
The test captures the intent of the flow. Spur figures out how to execute it against the application in front of it. This means the same test can keep working across platforms or ui changes, because the thing you’re testing - the user behavior - hasn’t changed.
Spur runs the infrastructure too
You upload the builds and choose what you want to validate. Spur handles the environment those tests execute in and runs them on our infrastructure. That means mobile testing becomes much closer to:
build → upload → validate
instead of becoming its own infrastructure project.
Tests can run as part of a release, against a pull request, on a schedule, or whenever an engineer wants to validate a change.
Built for engineers and agents
As more software is written by coding agents, we think validation also needs to become something agents can use directly.
MCP is a first-class interface to Spur. A coding agent can make a change, ask Spur to validate the relevant user flows, inspect the result, fix what failed, and run the validation again.
So the loop starts to look like:
generate → validate → fix → validate → ship
Humans can define the intent, using anything as context (tickets, figma files, PRDs, your codebase). Agents can continuously check whether the software still satisfies it.
We've spent the last few years building testing infrastructure for large software teams.
One thing has become increasingly clear: writing software is becoming dramatically easier, while proving that software works is still slow and expensive.
We think the gap between software generation and our ability to validate it is only going to widen.
The future of testing isn't maintaining more and more scripts. Validation needs to become more intelligent, adaptive, and be able to keep up with the rate at which software is changing.
That's what we're building at Spur.
If you're building or maintaining a mobile app, we'd love to work with you: https://spurtest.com/book-a-demo.
Anushka and I met our freshman year at Yale. We did research at the Yale NLP Lab on web agents. Sneha previously worked on growth at Figma, and Anushka on codegen at DeepMind. We’ve hacked on our fair share of projects before — and have very much felt the pain of neglecting testing.