pguso/ai-agents-from-scratch
Demystify AI agents by building them yourself. Local LLMs, no black boxes, real understanding of function calling, memory, and ReAct patterns.
This project helps software developers understand how AI agents work by building them step-by-step using local large language models (LLMs). Developers learn to combine an LLM with tools, memory, and reasoning patterns to create agents that can perform tasks. It takes fundamental concepts and outputs functional, explainable code for various agent architectures.
3,283 stars.
Use this if you are a software developer who wants to deeply understand the mechanics of AI agents before using complex frameworks, focusing on local LLMs and transparent design.
Not ideal if you are looking for a ready-to-use AI agent framework or a quick solution for an existing problem, as this project is purely educational and focused on foundational understanding.
Stars
3,283
Forks
488
Language
JavaScript
License
MIT
Category
Last pushed
Jan 21, 2026
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/agents/pguso/ai-agents-from-scratch"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Related agents
InfinitiBit/graphbit
GraphBit is the world’s first enterprise-grade Agentic AI framework, built on a Rust core with a...
autogluon/autogluon-assistant
Multi-Agent System Powered by LLMs for End-to-end Multimodal ML Automation
pguso/agents-from-scratch
Build AI agents from first principles using a local LLM - no frameworks, no cloud APIs, no...
samholt/L2MAC
🚀 The LLM Automatic Computer Framework: L2MAC
mnfst/awesome-free-llm-apis
Permanent Free LLM API List (API Keys) 😎🔑