agents-from-scratch
Build AI agents from first principles using a local LLM - no frameworks, no cloud APIs, no hidden reasoning.
- 1.1k
- Python
- MIT
- 4 mo ago
Build AI agents from first principles using a local LLM - no frameworks, no cloud APIs, no hidden reasoning.
Local-first AI teaching assistant for editable lesson drafts, student materials, and slides. Uses your curriculum and chosen model. Teacher-reviewed beta.
Complete Python curriculum: zero experience → full-stack mastery. 274 hands-on projects, 34 concept guides, 248 solution walkthroughs, AI-native tutoring, spaced repetition. Free & open source.
FLAMES is an interactive web-based training platform designed to teach firewall and network security concepts in a hands-on, gamified manner. It simulates a FortiOS-styled web interface where students configure firewall policies, network interfaces, and port assignments.
A hands-on, browser/terminal-based platform to learn Git by actually using it — not by reading about it.
A step-by-step, beginner-friendly implementation of Convolutional Neural Networks from scratch using the CIFAR-10 dataset, with visualizations, experiments, and detailed explanations.
Evil Lang | Educational language for exploring interpreters & closures A lightweight Python-implemented language blending functional/imperative paradigms
Chinese-first Bitcoin study quests: task-based learning for keys, UTXO, nodes, Script, multisig, Lightning, and privacy