LukeW Details His Personal AI Pipeline: Hourly Auto-Indexing of Audio, Video and PDFs
LukeW · x · 2026-10-20
Luke Wroblewski explains how he improved content indexing for Ask LukeW, his three-year-old personal AI Q&A feature, arguing a personal AI is only as good as its content.
Key practices: an ingestion pipeline that hourly checks for new/changed content and queues audio, video, PDFs and images for background processing; voice-aware transcription that consistently labels him in transcripts; splitting transcripts into passages with timestamps so answers link to exact moments in embedded players; a vision model that extracts text and describes charts in PDFs, with each slide saved separately for inline display; and image analysis that combines auto-descriptions with his original captions.
More from coding & agent
- Factory adds Linear delegations, letting Droid agents run issues automatically — matanSF · 2026-10-21
- Agent memory, goals and emotion should be designed as one system, argues dev — mimi10v3 · 2026-10-21
- Meta launches Muse Wonder coding models with 1M context, $4/$20 per million tokens — AIatMeta · 2026-10-21
- CastWorkshop: open-source studio ties Ollama and ComfyUI into persistent AI characters — CastWorkshop · 2026-10-21
- Agents break in production for boring backend reasons, not model reasons — DayKnown1050 · 2026-10-21
- REA turns any site, app, or game into an RL environment via open-source AgentEnv — henloitsjoyce · 2026-10-21