How daykalif/hello-agents works
Hello-Agents is an open, practice-oriented curriculum for understanding and building AI-native agents, rather than a single deployed agent product. The learner moves from agent and LLM fundamentals through agent patterns and framework construction, then into memory, context, protocols, training, and evaluation, before applying those ideas in larger projects. The graph preserves the README’s major learning branches; only the code paths present in the supplied tree or excerpts are attached to nodes. Other chapter and project implementations were not sampled, so their internal wiring is not asserted.
Main entry points: 10_RAG_Pipeline_Complete.py Memory and RAG · llm_client.py Classic Patterns · YYHDBL-HelloCodeAgentCli Framework Construction · 06_complete_pipeline.py Agentic Training
Components 20
Foundations
- Agent Basics
- Agent History
- LLM Fundamentals
Agent Building
- Classic Patterns
code/chapter4/llm_client.py - Low-Code Platforms
- Agent Frameworks
- Framework Construction
Co-creation-projects/YYHDBL-HelloCodeAgentCli
Advanced Systems
- Memory and RAG
code/chapter8/10_RAG_Pipeline_Complete.py - Document Ingestion
code/chapter8/04_RAGTool_MarkItDown_Pipeline.py - Context Engineering
- Agent Protocols
- Agentic Training
code/chapter11/06_complete_pipeline.py - Agent Evaluation
Applications
- Travel Assistant
- Deep Research
- Cyber Town
- Graduation Project
- Interview Prep
- Agent Builder
Other components
- Learner
Connections 21
- Learner → Agent Basics: studies
- Agent Basics → Agent History: continues to
- Agent History → LLM Fundamentals: continues to
- LLM Fundamentals → Classic Patterns: learns patterns
- Classic Patterns → Low-Code Platforms: explores platforms
- Low-Code Platforms → Agent Frameworks: explores frameworks
- Agent Frameworks → Framework Construction: builds framework
- Framework Construction → Memory and RAG: extends with memory
- Memory and RAG → Document Ingestion: practices ingestion
- Memory and RAG → Context Engineering: studies context
- Context Engineering → Agent Protocols: studies protocols
- Agent Protocols → Agentic Training: studies training
- Agentic Training → Agent Evaluation: evaluates agents
- Agent Evaluation → Travel Assistant: applies learning
- Agent Evaluation → Deep Research: applies learning
- Agent Evaluation → Cyber Town: applies learning
- Travel Assistant → Graduation Project: informs project
- Deep Research → Graduation Project: informs project
- Cyber Town → Graduation Project: informs project
- Learner → Interview Prep: uses supplement
- Graduation Project → Agent Builder: builds agent