Modular Data Infrastructure For AI With Covalent
summary
Covalent releases an open‑source TypeScript AI Agent SDK that launches stateless LLM agents, composes them into a Zero Employee Enterprise workflow with a router that ping‑pongs reasoning, and wires agents to external tools via schema‑driven function calls. Example: a GPT‑4‑mini reporting agent calls GoldRush for token, NFT and transaction data across ~100 blockchains. Quickstart uses `npx covalenthq-cli create-zee-app` to scaffold a ZEE app; dozen LLM providers are already pluggable.
AI-written summary · about
chapters (13)
- 0:00Introduction & Overview
- 0:46Zero Employee Enterprise (ZEE)
- 1:20Chain of Thought Models Explained
- 1:51Agent Autonomy & Collaboration
- 2:36AI Agent SDK Blog Post
- 3:53Exploring Documentation
- 4:28Building an LLM Agent
- 5:52Tools & GoldRush On-Chain Data
- 7:02ZE Workflow & Routing
- 8:18Open-Source Repo Tour
- 11:08Quickstart: Create Z App
- 12:10Use Cases & Contributing
- 13:17Final Thoughts & Hackathon Origins
related talks
48mAI Coding Agents and How to Code Them
34mAI Agents: Transforming Autonomous Decision-Making
19mBuilding Data-Driven, Intelligent Apps in the Age of GenAI
37mAGENTS OF S.E.A.L.E.D: AI Agentic Cybersecurity Framework
07h 47mAI Agents Conference 2026 - Day 5
50mOpening Keynote
28mBeyond Vibe Coding: AI Agents for Technical Debt Detection
01h 28mAI Coding Agents and How to Code Them