# PR reviews before deployment, autonomous repository onboarding, and Claude Haiku 5.5 > 3 updates: Repository onboarding runs as an autonomous analysis, Start PR reviews before the preview deployment is ready, A regression review model… Source: https://qa.tech/changelog/2026-10-26 --- 3 updates shipped on October 26, 2026. In this release 1. 01[Repository onboarding runs as an autonomous analysis](#2026-10-26-repository-onboarding-autonomous-analysis)Highlight 2. 02[Start PR reviews before the preview deployment is ready](#2026-10-26-start-pr-reviews-before-deployment)Highlight 3. 03[A regression review model setting and Claude Haiku 5.5](#2026-10-26-regression-review-model-and-haiku) 1. Highlight Featured this release ## Repository onboarding runs as an autonomous analysis After you connect a repository, QA.tech reviews your recent PRs and test suite, then fills the gaps it finds. A chat agent samples 10–12 recent PRs and classifies how ready QA.tech was to test each one, maps your existing tests when code access is on, and closes gaps with PR review rules, memories, todos, targeted crawls and new tests. In the repository settings, Scan is now Analyse and opens the analysis conversation so you can follow along. More on this: [integrations](https://qa.tech/product/integrations), [AI test generation](https://qa.tech/use-cases/ai-test-generation) and [agentic QA tools](https://qa.tech/blog/best-agentic-qa-tools-in-2026). [\# Direct link](#2026-10-26-repository-onboarding-autonomous-analysis) 2. Highlight Featured this release ## Start PR reviews before the preview deployment is ready QA.tech can begin reviewing a pull request as soon as it opens, then run tests once your preview lands. Turn on Start Before Deployment (beta) under Settings → Integrations → GitHub App when editing a repository. The review researches the change and plans tests while your preview builds, the PR comment shows "Waiting for the preview deployment", and tests run only against the deployment for that commit. Review time in the stats counts from when the deployment was ready. ![The assistant planning PR tests while the preview deployment is still building](https://qa.tech/changelog-assets/2026-10-26-start-before-deployment.webp) More on this: [PR testing](https://qa.tech/product/pr-testing) and [PR and CI/CD testing](https://qa.tech/use-cases/pr-cicd-testing). [\# Direct link](#2026-10-26-start-pr-reviews-before-deployment) 3. Update 03 of 03 ## A regression review model setting and Claude Haiku 5.5 Pick which model reviews your test plan runs, and use Haiku 5.5 for faster chat turns. Project → Settings → Models has a new Regression review row, with the same picker as Chat Assistant and Pull Request. It applies to reviews from Run Tests, schedules, CI and the API. ![Haiku 5.5 in the chat composer's model menu](https://qa.tech/changelog-assets/2026-10-26-haiku-5-5.webp) In the chat composer, turn Auto off and pick Haiku 5.5 next to Opus 5.5 and Sonnet 5.5. It has a 1M-token context window and adaptive thinking. More on this: [automated regression testing](https://qa.tech/use-cases/automated-regression-testing) and [product](https://qa.tech/product). [\# Direct link](#2026-10-26-regression-review-model-and-haiku) ## Your team moves fast. Can your testing keep up? QA.tech agents test your product autonomously, so moving fast never means shipping broken. See how it works in a 30‑minute demo. [Book a demo](https://qa.tech/demo) [QA.tech](https://qa.tech/)[SOC 2 Type 2 compliant![Sensiba SOC 2 Type 2 compliant](https://qa.tech/assets/soc2-type2-badge-DPb1j9VX.png)](https://trust.qa.tech/) ### Product - [API Testing](https://qa.tech/product/api-testing) - [MCP](https://qa.tech/product/mcp) - [Mobile Testing](https://qa.tech/product/mobile-testing) - [PR Testing](https://qa.tech/product/pr-testing) - [Voice Testing](https://qa.tech/product/voice-testing) - [Web Testing](https://qa.tech/product/web-testing) - [AI QA testing guide](https://qa.tech/ai-qa-testing) - [Changelog](https://qa.tech/changelog) - [Integrations](https://qa.tech/product/integrations) - [Platform overview](https://qa.tech/product) - [Use cases](https://qa.tech/use-cases) ### Compare - [QA.tech vs Cypress](https://qa.tech/compare/qa-tech-vs-cypress) - [QA.tech vs Mabl](https://qa.tech/compare/qa-tech-vs-mabl) - [QA.tech vs Playwright](https://qa.tech/compare/qa-tech-vs-playwright) - [QA.tech vs QA Wolf](https://qa.tech/compare/qa-tech-vs-qa-wolf) - [B2B SaaS testing](https://qa.tech/saas) - [Ecommerce testing](https://qa.tech/industries/ecommerce) - [For CTOs](https://qa.tech/solutions/ctos) - [For QA teams](https://qa.tech/solutions/qa-testers) ### Resources - [Docs](https://docs.qa.tech/) - [Blog](https://qa.tech/blog) - [Events](https://qa.tech/events) - [Pricing](https://qa.tech/pricing) - [Case Studies](https://qa.tech/case-studies) ### Company - [About](https://qa.tech/about-us) - [Partner Program](https://qa.tech/partnerships) - [Press](https://qa.tech/press) - [Contact Us](https://qa.tech/contact) - [Privacy Policy](https://qa.tech/privacy-policy) - [Terms of Service](https://trust.qa.tech/resources?s=mb9w4yjb03i3t8bjsq4jx3&name=main-agreement.pdf) - [Trust Center: Security & Compliance](https://trust.qa.tech/) ### Social - [](https://www.linkedin.com/company/qa-tech-ai/) - [](https://x.com/getqatech) - [](https://www.instagram.com/getqatech/) © 2026 QA.tech. All rights reserved