Skip to content

agentic-artificial-intelligence

PhD Course: Agentic Artificial Intelligence - LLM Agents, RAG, Multi-Agent Systems

View on GitHub


Publications

ReAct: Synergizing Reasoning and Acting in Language Models

While large language models (LLMs) have demonstrated impressive capabilities across tasks in language understanding and interactive decision making, their abilities for reasoning (e.g. chain-of-thought prompting) and acting (e.g. action plan generation) have primarily been studied as separate topics...

Property Value

| arXiv | 2210.03629 |

| Year | 2022 |

Authors: Shunyu Yao, Jeffrey Zhao, Dian Yu, Nan Du, Izhak Shafran, Karthik Narasimhan, Yuan Cao


Information

Property Value
Language Python
Stars 0
Forks 0
Watchers 0
Open Issues 1
License No License
Created 2025-12-24
Last Updated 2026-02-19
Last Push 2026-01-04
Contributors 1
Default Branch main
Visibility private

Notebooks

This repository contains 14 notebook(s):

Notebook Language Type

| L01_first_agent | PYTHON | jupyter |

| L02_prompting_strategies | PYTHON | jupyter |

| L03_function_calling_comparison | PYTHON | jupyter |

| L03_mcp_tool_implementation | PYTHON | jupyter |

| L04_reflexion_implementation | PYTHON | jupyter |

| L05_coordination_demo | PYTHON | jupyter |

| L05_message_passing | PYTHON | jupyter |

| L06_LangGraph_Agent | PYTHON | jupyter |

| L07_Self_RAG | PYTHON | jupyter |

| L08_GraphRAG | PYTHON | jupyter |

| L09_Verification | PYTHON | jupyter |

| L10_Benchmarking | PYTHON | jupyter |

| L11_Code_Agent | PYTHON | jupyter |

| L12_Generative_Agents | PYTHON | jupyter |

Datasets

This repository includes 13 dataset(s):

Dataset Format Size

| .lighthouserc.json | .json | 0.63 KB |

| datasets | | 0.0 KB |

| .gitkeep | | 0.0 KB |

| glossary_inventory.json | .json | 30.6 KB |

| term_index.json | .json | 19.18 KB |

| charts.json | .json | 4.45 KB |

| course.json | .json | 2.36 KB |

| exercises.json | .json | 4.25 KB |

| glossary.json | .json | 0.31 KB |

| quizzes.json | .json | 0.44 KB |

| weeks.json | .json | 0.6 KB |

| week_fixes.json | .json | 4.6 KB |

| quality_report.json | .json | 2.66 KB |

Reproducibility

This repository includes reproducibility tools:

  • Python requirements.txt

Status

  • Issues: Enabled
  • Wiki: Enabled
  • Pages: Enabled

README

Agentic Artificial Intelligence

PhD-level course on LLM-based autonomous agents, multi-agent systems, and advanced RAG architectures.

Course Overview

This 12-week course covers the theory and practice of building intelligent agents powered by large language models. Students will learn to design, implement, and evaluate autonomous systems that can reason, plan, and act in complex environments.

Prerequisites

  • Machine Learning fundamentals
  • Python proficiency (3.10+)
  • Experience with LLM APIs (OpenAI, Anthropic)
  • Basic knowledge of transformers and attention mechanisms

Course Structure

Module Weeks Topics
Foundations 1-2 Agent definition, ReAct, LLM prompting, CoT/ToT
Single-Agent Systems 3-4 Tool use (MCP), planning, memory, Reflexion
Multi-Agent Systems 5-6 Architectures, LangGraph, AutoGen, CrewAI
RAG & Knowledge 7-8 Advanced RAG, GraphRAG, knowledge graphs
Safety & Evaluation 9-10 Hallucination prevention, benchmarks, red-teaming
Applications 11-12 Domain agents, research frontiers, projects

Getting Started

  1. Clone the repository
  2. Install dependencies: pip install -r requirements.txt
  3. Set up API keys (OpenAI, Anthropic, etc.)
  4. Navigate to lesson folders for materials

Repository Structure

agentic-artificial-intelligence/
|-- L01_Introduction_Agentic_AI/
|   |-- L01_Introduction_Agentic_AI.tex
|   |-- 01_agent_definition/
|   |   |-- chart.py
|   |   |-- chart.pdf
|   |-- notebooks/
|   |-- exercises/
|   |-- readings/
|-- L02_LLM_Foundations_Agents/
|-- ...
|-- L12_Research_Frontiers/
|-- tools/
|   |-- hallucination_checks/
|   |-- quality/
|-- datasets/
|-- SYLLABUS.md
|-- PROGRESS_TRACKER.md
|-- requirements.txt

Tools & Frameworks

  • LangGraph: State machine-based agent framework
  • AutoGen: Conversation-based multi-agent
  • CrewAI: Role-based team agents
  • MCP: Model Context Protocol for tool use

Evaluation

Component Weight
Weekly Exercises 30%
Midterm Project (Week 6) 20%
Paper Presentation 15%
Final Project 35%

Resources

Contributing

See SYLLABUS.md for course schedule and PROGRESS_TRACKER.md for development status.

License

Educational use only. Contact instructor for permissions.