🤖 Autonomous Multi-Agent AI System with Memory

Agentic AI Developer Certification – Module 3 Project

---

📘 Overview

This project implements an Autonomous Multi-Agent System with Memory, where agents can plan, execute, evaluate, and recall tasks. Developed as part of the Agentic AI Developer Certification, it demonstrates how an AI system can reason over time, reuse past goals, and improve autonomously.

---

🧩 System Components

The system supports memory recall: on subsequent runs, it offers to resume previous goals from memory.

---

🧠 Example Run

🤖 Autonomous Multi-Agent System with Memory is running...

🧠 Previous goal found in memory: how to become a full stack developer
Would you like to continue from last goal? (y/n): y

[Planner] Creating plan...
Plan:
  - 1: web_search -> find relevant sources about full-stack development
  - 2: analyze -> extract key technologies and steps
  - 3: summarize -> produce a concise roadmap summary

[Executor] Running tasks...
Result for task 1:
Found 3 sources: Paper A, Blog B, Roadmap C covering React, Node.js, Databases

Result for task 2:
Analysis: Full-stack developers combine front-end (React, CSS) and back-end (Node.js, Express, DB) skills; emphasis on building projects.

Result for task 3:
Summary: To become a full-stack developer, master front-end frameworks (React/Angular), back-end frameworks (Node.js/Express, Python/Django), and databases (SQL/NoSQL). Build full projects, deploy them, and document your learning.

[Evaluator] Reviewing outputs…
Review result:
The roadmap is concise and actionable but could include more project examples. Score: 85/100.

✅ Run complete. All outputs stored in `memory/memory_store.json`
🏗️ Architecture & Folder Structure
graphql
Copy code
autonomous_ai_project/
│
├── main_autonomous.py        # Orchestrates workflow
├── llm_client.py             # Handles LLM communication (Groq or stub)
│
├── agents/
│   ├── __init__.py
│   ├── planner.py            # Creates plans
│   ├── executor.py           # Executes tasks
│   ├── evaluator.py          # Reviews and scores outputs
│   └── memory_agent.py       # Handles persistent memory
│
├── memory/
│   └── memory_store.json     # Stores history of goals/tasks/results
│
├── .env.example              # Placeholder for API key
├── .gitignore                # Excludes .env, .venv, cache
└── requirements.txt          # Dependencies: python-dotenv, groq (optional)
✅ Key Features
Autonomous multi-step reasoning and task execution

Persistent memory enabling context recall across sessions

Modular architected agents enabling clear separation of concerns

Integration with Groq Llama 3 model (or offline stub mode)

Secure handling of API key via .env and .gitignore

💡 Real-World Applications
Use Case	Example
Automated Research	System that plans web search, extracts and summarizes papers
Software Task Planner	AI that breaks down project goals into tasks, executes code checks
Learning Assistant	Tracks previous learning goals, suggests next topics, remembers progress
Business Intelligence	Remembers past reports, automates recurring analytics and summaries

🧪 Setup & Run Instructions
Clone repository:

bash
Copy code
git clone https://github.com/shaiknawaz568/autonomous-ai-project.git
cd autonomous-ai-project
Create virtual environment:

bash
Copy code
python -m venv .venv
.\.venv\Scripts\Activate.ps1   # On Windows PowerShell
Install dependencies:

bash
Copy code
pip install -r requirements.txt
Add your API key:
Create .env with:

ini
Copy code
GROQ_API_KEY=your_api_key_here
If no key is provided, system works in stub (offline) mode.

Run system:

bash
Copy code
python main_autonomous.py
🏁 Certification Context
This project fulfills Module 3 of the Agentic AI Developer Certification (AAIDC) by Ready Tensor, focused on Autonomous Systems with Memory.

Module	Focus	Project
1️⃣	Single-Agent System	Intelligent AI Assistant
2️⃣	Multi-Agent System	Multi-Agent Reasoning
3️⃣	Autonomous System with Memory	This project — Autonomous Multi-Agent System

👨‍💻 Author
Shaik Nawaz Shareef
Agentic AI Developer | Data Analyst | Full Stack Enthusiast
GitHub: https://github.com/shaiknawaz568