Vibeleaderboard
← All Intel
Intel / video

500 Skills, Zero Fine-Tuning: LinkedIn's Playbook for AI Agents — Ajay Prakash, LinkedIn

Source
youtube.com
Author
AI Engineer
Date

Context

LinkedIn's coding- system makes a large tool collection available through a small interface: discover a relevant tool, inspect its required inputs, then execute it. The conference presentation describes this approach to avoiding a crowded with every available tool.

The company's engineering account explains that playbooks also appear as callable tools rather than only prompt templates. They supply the steps and references needed for internal workflows, so agents can retrieve organizational instructions when needed. That article has its own earlier Intel entry.

Terms in this piece · Glossary
  • AI agent — An AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.
  • context window — The maximum amount of text a model can consider at once — its working memory for the current conversation or task.
  • MCP — The Model Context Protocol — an open standard that lets any AI assistant plug into any tool or data source without custom integration code.
Chapters
  • 0:00An on call incident, handled end to end
  • 2:48Why coding agents failed inside LinkedIn
  • 5:10A thousand repos on internal frameworks
  • 6:34An internal MCP, starting with code search
  • 8:30Why tools alone were not enough
  • 10:34Playbooks, instructions served as tools
  • 12:27Two rules, self contained and broken up
  • 14:23The loop that keeps playbooks fresh
  • 15:34Architecture, local and central playbooks
  • 17:23Three meta tools to scale past the MCP limit
  • 19:01Takeaways
Read the source www.youtube.com
More from AI Engineer
Recommended reads
Comments

Checking sign-in…

Loading comments…