Multi-Agent Loop Workflow Curates 1,132 Top Bilibili Videos Into a Website
huangyun_122 · x · 2026-11-04
Inspired by Karpathy's HackerTube, a developer built a site called "AnliBili" using Teamily AI's team agents: it combed through roughly 10 million discussions on Zhihu and Xiaohongshu spanning nearly 16 years, and distilled 1,132 high-quality Bilibili videos that were each recommended by at least 3 people across 3 different months.
The workflow runs as a continuously executing Loop: once set up, the agent team keeps cycling through "organize comments → filter videos → categorize and publish," with almost no human intervention. The author and all agents share context in the same group chat—just give feedback like "add Xiaohongshu as a data source," and the agents automatically reassign tasks, adjust, and rerun the whole loop to update the site.
More from coding & agent
- Revolut cuts AI coding costs 10x by dropping Claude for self-hosted models — beasthunterr69 · 2026-11-04
- Fitting the OODA Loop to AI Agents: A Practical Framework for Iterative Optimization — DanielLockyer · 2026-11-04
- Mistral Large 4 tested on real Django CVE: 133 agents find, prove and patch the bug — MistralAI · 2026-11-04
- Dev calls to phase out C++ as tangled mess; formal specs key to proving AI code correct — thomasahle · 2026-11-04
- SOMA lets Claude Code do CAD: 1,000+ users, full bookshelf-to-workshop-drawing tutorial — Lazar988 · 2026-11-04
- We built MCP, API and CLI integrations for AI agents — here's how they compare — hottown · 2026-11-04