AI Resume Grader — CS AI Workshop Starter

You already have an application. Your job is to give it AI.

The Gradio frontend is finished. During the workshop, you will connect the app to Google Gemini in ai.py, then experiment with the grading prompt in prompt.py.

What you need

• Python 3.9 or newer
• A Gemini API key from Google AI Studio: https://aistudio.google.com/api-keys

1. Open a terminal in this folder

macOS / Linux

python3 -m venv .venv
source .venv/bin/activate
python -m pip install -r requirements.txt

Windows PowerShell

py -m venv .venv
.venv\Scripts\Activate.ps1
python -m pip install -r requirements.txt

2. Add your API key

Copy .env.example to a new file named .env.

macOS / Linux:

cp .env.example .env

Then put your own key in .env:

GEMINI_API_KEY=your_actual_key_here

Never post your key in Discord or push it to GitHub.

3. Run the starter app

python app.py

Open the local Gradio URL printed in your terminal. The interface should run, but clicking Grade Resume will say the AI is not implemented yet. That is intentional.

4. Your task

Open ai.py and work through the TODOs:

1. Read GEMINI_API_KEY from the environment.
2. Import genai from Google and create a Gemini client.
3. Import and call build_grading_prompt from prompt.py.
4. Send the prompt to Gemini.
5. Return Gemini's text response to the frontend.

Then open prompt.py and make the grader your own. Change the rubric, tone, categories, or requested feedback and see how the output changes.

Project flow

Resume text
   ↓
Gradio frontend (`app.py`)
   ↓
Python function (`ai.py`)
   ↓
Gemini API
   ↓
Gemini model
   ↓
AI response
   ↓
Gradio displays feedback

Test resumes

Use any file inside sample_resumes/ to test your grader.

Optional extensions

• Compare a resume to a job description
• Add category scores
• Add PDF upload
• Save past results
• Create a job-match analyzer
• Later connect outside tools/data through MCP