Dev LLM-indexes 12 years of 500K tweets into a private searchable archive with photo analysis
Yamapama · x · 2026-11-06
Dutch developer @Yamapama built a personal archive because X's search is unusable: he pushed 500K tweets spanning 12 years through an LLM and wrapped them in a custom UI with timeline search, year filters (2011–2026), topic clusters, mood/keyword signals, and geolocation filters — 590,537 tweets indexed.
The standout is multimodal processing: every photo and video attached to tweets is analyzed too, generating Dutch-language descriptions, keyword tags, metadata, and mood signals — e.g. a 2014 photo of a homemade liquor bottle label gets tagged with details like "product shot," "lipstick kisses," and full metadata from the original Twitter archive export. He's still fixing the UI and jokes he yells at his LLM when a description comes out wrong.
More from coding & agent
- Perplexity Computer builds 6-minute GTA RAGE engine explainer from 40 web sources — jeffgrimes9 · 2026-11-06
- Agents still claim tasks will take 'several hours' — then finish in under 10 minutes — jacalulu · 2026-11-06
- Alchemy teases next-week release: Rivet and Effect Cluster as swappable deploy layers — samgoodwin89 · 2026-11-06
- Grok Build v1.0.75 ships major reliability and performance upgrades for multi-agent workflows — elonmusk · 2026-11-06
- Gauge open-sources CLI as 85% of coding sessions are now agent-initiated — ycombinator · 2026-11-06
- kaigen3d: a C-based agent-native game engine with headless operation and automated QA — gdechichi · 2026-11-06