Releases · hoangsonww/DocuThinker-AI-App

🚀 v2.1.2 – Stable Public Release

DocuThinker is now officially production-ready, publicly usable, and rock-solid stable.

Performance & Reliability
- Eliminated cold-starts by keeping the backend “warm” on Render/Vercel
- Redis caching for faster repeat requests
- Rate-limit protection on all AI calls
- Increased concurrency with RabbitMQ workers
UI/UX Improvements
- Persistent light/dark mode toggle
- Drag-and-drop upload with progress bars
- Friendly error screens with retry buttons
Enhanced AI Features
- Improved voice-to-text chat accuracy
- Bullet-point, multi-language summaries
- Paragraph-level sentiment analysis
- Contextual content rewriting and recommendations
API & GraphQL Updates
- Cursor-based pagination on document listings
- Bulk delete and theme-update GraphQL mutations
- Unified JSON error schema (code + details)
DevOps & CI/CD
- Updated Docker Compose and Kubernetes manifests for v2.1.2
- Jenkins and GitHub Actions now run full integration tests (MongoDB + Redis)
- Helm chart for one-click cluster deployment
- Prometheus health checks enabled on all services
Mobile Beta
- Expo-based React Native app now feature-complete for upload, chat, and browsing
- Launch locally with npx expo start

Checkout tag:
```
git fetch --tags
git checkout v2.1.2
```
Install & build:
```
npm ci
npm run build
```
Configure your .env with keys for Firebase, Google AI, Redis, RabbitMQ, etc.

Run locally:

aicc dev
cd mobile-app && npx expo start

If you encounter any issues or have feature requests, please open an issue in the repository. Thank you for using DocuThinker!