We got tired of spending half our sprints writing test cases manually. softAssert exists to give QA teams back their time — so they can focus on finding real bugs, not writing boilerplate.
Our mission
Test cases are a means to an end: finding bugs before users do. When writing them takes 4 hours per sprint, something is broken. softAssert uses AI to handle the structure so your team can focus on the thinking that actually matters.
How it works
No setup, no training data, no complicated config.
Paste a user story, PRD excerpt, or plain English description of a feature. softAssert doesn't need structured input — natural language works fine.
Claude AI analyzes your description and generates a structured set of test cases: happy paths, negative tests, edge cases, and security checks.
Run suites directly in softAssert, mark pass/fail, track trends. Export to Jira, Linear, Excel, or Playwright test skeletons.
softAssert learns what kinds of tests matter for your product. Bug history informs test focus. Coverage gaps become visible.
Our story
We were QA engineers ourselves, spending hours writing test cases for every sprint. There had to be a better way.
Built the first AI test generation prototype using Claude. Generated 40 test cases in 45 seconds. We knew we were onto something.
Opened the beta to 50 QA engineers. Feedback poured in — bug reporting, regression suites, and Jira integration were the most requested features.
softAssert v1.0 launched publicly with AI generation, bug tracking, suite execution, and full export. 500+ sign-ups in the first month.
Our values
We build for QA engineers, not managers. Every feature is designed around the workflow of someone who actually writes and runs tests.
AI accelerates the tedious parts — not the thinking. We generate structure so you can focus on what matters: finding real bugs.
Your test cases and bug reports are yours. We never use your project data to train AI models, full stop.
QA tooling doesn't have to be complicated. We obsess over removing friction so you spend less time clicking and more time testing.
Good test cases follow patterns. We encode QA best practices into every generation — happy paths, edge cases, security checks.
Features come from real conversations with QA engineers. If you have a workflow we should support, we want to hear about it.