-
A Multi-Agent Interview Preparation System built using Google ADK and Gemini 2.5 Flash.
AI Interview Coach
A Multi-Agent Interview Preparation System built using Google ADK and Gemini 2.5 Flash.
Live Demo
https://interview-coach-agent.onrender.com
Features
- Java Interview Agent
- SQL Interview Agent
- DSA Interview Agent
- Answer Evaluation Agent
- Personalized Study Plan Agent
- Coordinator Agent for intelligent routing
Architecture
The Coordinator Agent analyzes user requests and routes them to the appropriate specialist agent.
User → Coordinator Agent → Specialist Agent
Specialist Agents:
- Java Agent
- SQL Agent
- DSA Agent
- Evaluation Agent
- Study Plan Agent
Tech Stack
- Python
- Google ADK (Agent Development Kit)
- Gemini 2.5 Flash
- Multi-Agent Architecture
- GitHub
- Render
Example Usage
Java Interview
User: Start Java Interview
Agent: Explain the difference between HashMap and ConcurrentHashMap.
SQL Interview
User: Start SQL Interview
Agent: Explain the difference between INNER JOIN and LEFT JOIN.
DSA Interview
User: Start DSA Interview
Agent: Given an array of integers, find two numbers that add up to a target value.
Run Locally
Install dependencies:
pip install -r requirements.txt
Start the application:
adk web
Project Structure
interview-coach-agent/
│
├── my_agent/
│ ├── agent.py
│ ├── coordinator_agent.py
│ ├── java_agent.py
│ ├── sql_agent.py
│ ├── dsa_agent.py
│ ├── evaluation_agent.py
│ └── studyPlan_agent.py
│
├── requirements.txt
├── README.md
└── .gitignore
Known Limitation
This project currently uses the Gemini API free tier.
If the API quota is exhausted, the application may temporarily be unable to generate responses until the quota resets.
Future Improvements
- Interview score tracking
- Database integration
- User authentication
- Interview history
- Performance analytics dashboard
- Custom scoring engine
Author
Laxmi Shekhappa Bhadrapura
