QA.tech
Autonomous AI QA agents that explore and test your app like real users.
About
QA.tech replaces scripted tests with autonomous AI agents that explore your application the way customers do. You describe scenarios in plain English or let the agents roam, and they navigate complex flows including third-party integrations, email verification, and mobile-to-web handoffs, filing developer-friendly bug reports with logs. The agents run in CI/CD and review pull requests, and recent releases added API testing, voice-flow testing, and OAuth sign-in for MCP clients like Cursor and Claude Code, so coding agents can drive the QA loop.
The Stockholm startup was founded in 2023 and raised a 3M euro seed round in 2024 led by byFounders, PROfounders, and Curiosity VC, and it ships as a SaaS subscription. The agentic approach trades determinism for coverage: execution is AI-interpreted at runtime, so teams review agent outcomes rather than fixed assertions. For teams drowning in flaky E2E suites, that tradeoff is the whole appeal; for regulated release gates, it needs a human sign-off layer.
What it can do
- Autonomous AI agents exploring apps like real users
- Plain-English test scenario authoring
- API testing as part of agent runs
- Voice and dictation flow testing
- MCP client support with OAuth (Cursor, Claude Code)
- PR reviews with before/after bug reproduction
- Exploratory edge-case and UX inconsistency detection
Setup
- Connect your repository and CI/CD pipeline to QA.tech.
- Describe test scenarios in plain English or let agents explore.
- Review generated bug reports and adjust scenarios.
- Enable PR reviews and scheduled agent runs.
Prerequisites
Web or mobile application; CI/CD integration for automated runs.
Use cases
Works with
Trust score — 7/10
Scored 7/10 (TIER verified, metrics-v1). Evidence: founded 2023 in Stockholm; 3M euro seed raised May 2024 (byFounders, PROfounders, Curiosity VC) documented by investor and trackers; API/voice/MCP features in current changelog; customer case study (Pricer) published. Strengths: well-funded seed startup, agentic model is genuinely differentiated, active feature velocity. Docked for: (1) young company with small customer base; 95%-of-bugs claims are vendor-reported; (2) non-deterministic execution makes release gating harder without human review. License: proprietary.
How the trust score is built
Full methodology →- Adoption · stars/downloads, log-scaled vs peers
- 6.0/10
- Maintenance · recency, releases, issue load
- 8.0/10
- Openness · license + docs
- 5.0/10
- Manual review · sanity read + red-flag check
- 7.0/10
- · Founded 2023 in Stockholm; 3M euro seed raised May 2024 (byFounders, PROfounders, Curiosity VC)
- · Agentic model genuinely differentiated; API/voice/MCP features in current changelog
- · Customer case study (Pricer) published; 3,664 LinkedIn followers
- · Young company with small customer base; 95%-of-bugs claims are vendor-reported
- · Non-deterministic execution makes release gating harder without human review
Sources
Every fact above was checked against these sources. Where a claim comes from the provider and couldn't be independently verified, the text says so.