Roadmap

Archive Infrastructure

S3 storage for archives, and API to serve it
Archive filtering and privacy settings
Postgres DB for pooled archives, and API to serve it
Row-level security
Small fixes:
- Split archive.json into files and chunks for convenient storage
- Download and store avatars
- Move from multiple archive_uploads to a single one
Implement tests for uploading and privacy
Embeddings infrastructure:
- Create and store embeddings
  - using people's openai api key or letting them pay us to do it
- Develop economics model for user charging
Browser extension for DB updates via Twitter client API scraping
Own your social graph: prompt users for contact info (email is included in archive), and display it in profile only to their mutuals.

Website
- Twitter-like advanced full text search
- User pages with top tweets and a button to copy as text (to put in LLM context)
Keyword and concept trends analysis
Website improvements:
- Restyle front page
- Render threads
Birdseye: personal archive insights (topic clustering app)
Train the tpot LLM and give it a twitter account
Semantic search
User summary pages (interests, projects, relationships)
Discourse mapping
Discussion reranking with bridging algorithms
GraphRAG on whole db
Interactions network visualization and temporal analysis