šŸ”„ Only 100 spots left for the 72% Lifetime Discount! Claim Now

v1.1.0 Update

Stability & Performance Update

What’s New?

We’ve been working hard to make Cubrain even more reliable and faster. This update focuses on performance optimization and a more engaging processing experience.

šŸš€ Performance & Optimization

  • Batch Processing: We now group 5-10 highlights into a single AI prompt. This significantly reduces API costs and speeds up the generation process by up to 60%.
  • Async Queue System: A new background processing system ensures that large PDFs don’t time out. You can now track your progress in real-time.
  • Empty File Validation: Added strict validation for file uploads. No more ā€œghostā€ decks from empty or unreadable PDFs!

šŸƒ Visual Progress UX (The Pacer)

We’ve introduced ā€œThe Pacerā€ā€”a real-time character animation that reflects the processing state.

  • Running: When AI is actively synthesizing your cards.
  • Success: When your deck is ready for review.
  • Purpose: We’ve turned technical latency into an entertaining visual element to keep you engaged while you wait.

šŸ›”ļø Stability & Subscription

  • Grace Period Logic: Pro users now have a 3-day grace period after their subscription expires before losing access to premium features.
  • Supabase Auth Refactor: Improved session management for a more stable login experience.

šŸ› Bug Fixes

  • Fixed a layout shift on the landing page when the launch banner was active.
  • Improved the real-time counter for the early bird discount.

šŸ› ļø Under the Hood: Batching & The Pacer

To handle the massive textbooks used by Med and Law students, we implemented a Vectorized Batching strategy. Instead of processing highlights one-by-one (which is slow and expensive), we group related annotations into a single high-context prompt.

Additionally, we’ve decoupled the AI processing from the web request using an Asynchronous Task Queue. This is why you see ā€œThe Pacerā€ animation—it’s not just a decoration, but a real-time listener to our backend queue, ensuring that even if a synthesis takes 30 seconds, your browser connection remains rock-solid.