Manufacturing AI Ecosystem 🏭

A unified manufacturing intelligence platform integrating multimodal GenAI for concept design, Agentic AI for supplier research, and a fully automated DevOps pipeline to AWS EKS.

---

🎯 What it does

Tab 1 — Multimodal GenAI Creator Enter a manufacturing concept to instantly generate a structured technical narrative (powered by Groq's lightning-fast Llama 3) alongside an AI-generated visual prototype (powered by Flux via Pollinations.ai).

Tab 2 — Agentic Sourcing System A two-agent CrewAI pipeline (Researcher + Writer) that takes a sourcing query, finds and scores real-world suppliers, and produces a ranked markdown report alongside a structured JSON artifact.

---

🛠️ Tech Stack

Layer	Technology
Frontend	Streamlit
LLM Engine	Google Gemini 2.5 Flash, Groq Llama 3.3 70B
Image Generation	Flux (via Pollinations.ai)
Agent Framework	CrewAI
Containerization	Docker
Orchestration	Kubernetes (AWS EKS)
CI/CD	GitHub Actions

---

📁 Project Structure

Manufacturing-AI-Ecosystem/
├── frontend/
│   └── app.py
├── backend/
│   ├── agents.py
│   ├── tasks.py
│   ├── orchestrator.py
│   ├── schemas.py
│   └── storage.py
├── k8s/
│   ├── deployment.yaml
│   └── service.yaml
├── .github/
│   └── workflows/
│       └── deploy.yml
├── artifacts/
├── Dockerfile
├── requirements.txt
└── eks-setup.ps1

---

💻 Running Locally

1. Clone the repository:

   git clone https://github.com/omtiwari17/Manufacturing-AI-Ecosystem.git
   cd Manufacturing-AI-Ecosystem

2. Install dependencies:

   pip install -r requirements.txt

3. Launch the application:

   python -m streamlit run frontend/app.py

4. Add your API Keys: When the UI loads in your browser, enter your keys in the sidebar:

   • Gemini API Key: Required for Tab 1 (Get it free at aistudio.google.com)
   • Groq API Key: Required for Tab 1 & Tab 2 (Get it free at console.groq.com)

---

🚀 AWS EKS Deployment

This application is configured for fully automated, zero-downtime deployments to Amazon Elastic Kubernetes Service (EKS).

1. Provision the Cluster (One-Time Setup)

Ensure you have the AWS CLI installed and configured (aws configure). Then, run the provisioning script to build the cluster and configure the necessary IAM roles:

.\eks-setup.ps1

2. Configure GitHub Secrets

Add the following repository secrets in GitHub (Settings -> Secrets and variables -> Actions):

• DOCKER_USERNAME (Your Docker Hub username)
• DOCKER_PASSWORD (Your Docker Hub Personal Access Token)
• AWS_ACCESS_KEY_ID (From AWS IAM)
• AWS_SECRET_ACCESS_KEY (From AWS IAM)
• AWS_REGION (e.g., ap-south-1)

3. Trigger Deployment

Any code pushed to the main branch will automatically trigger the GitHub Actions pipeline. The pipeline will build a new Docker image, push it to Docker Hub, and execute a rolling update on the EKS cluster.

git add .
git commit -m "Deploying unified ecosystem"
git push origin main

---

👥 Group Members

Sr No	Name	Enrollment Number
01	Om Tiwari	EN22CS301669
02	Paridhi Shirwalkar	EN22CS301684
03	Nitesh Chourasiya	EN22CS301660
04	Mradul Jain	EN22CS301616