Our strategic timeline to scale TubeRAG into the ultimate study companion for engineers, students, and workspaces. Join us on this journey!
Asynchronous processing queue powered by serverless Upstash QStash, returning instant 202 responses.
Chunking and embedding transcript text structures for semantic vector-aware querying and citations.
Real-time generation of structured outlines, concepts mindmaps, and cursive calligraphy study notes.
Secure user registration and token-based sessions utilizing Supabase Auth and JWT state verification.
Database layer isolation to secure user nodes, ensuring users can only read and write their own ingested workspaces.
Saving multi-agent chat sessions to the database to allow users to pause, resume, and manage historic queries.
Allowing users to organize ingested videos into folders and share workspaces or playlists with teammates.
Periodic frame extraction using vision models to transcribe texts from slide decks, code editors, and diagrams.
Inline voice tags support ([laughs], [whispers]) inside generated podcasts to create human-like audio conversations.
Downloading and printing high-definition interactive mindmaps as editable SVG vector assets.
Secure API keys and webhooks subscription endpoints to build integrations with external learning management tools.
Exporting semantic lecture transcripts in multiple formats including SRT, VTT, and styled Markdown reports.
Global semantic search index crossing transcripts, notes, and visual frames across all user repositories.