Placementor is an advanced interview preparation platform designed to assist users in practicing and refining their interview skills. Developed during a hackathon, this project utilizes sophisticated AI agents to simulate realistic interview scenarios and provide actionable feedback.
Demo: Youtube Video
- Resume Upload and Parsing: Users upload their resumes, which are parsed to extract relevant information.
- Interview Simulation: Users select a company, role, and interview round to generate a tailored set of interview questions.
- Speech Recognition: Users practice answering questions using voice input.
- Feedback and Evaluation: AI agents evaluate user responses and provide detailed feedback, including scores and improvement suggestions.
- Frontend: React.js, Tailwind CSS, Axios
- Backend: Python, Flask
- AI and NLP: Agno Framework, Gemini, SpeechRecognition, DuckDuckGo(Webscraping)
- Data Handling: JSON & localstorage
- Version Control: Git
- Authentication: Clerk
- Purpose: Extracts structured information from a user's uploaded resume.
- Logic:
- Uses the Gemini model within the Agno framework to parse the resume and extract key details such as full name, email, phone, skills, projects, education, and work experience.
- Constructs a prompt that instructs the AI to return the extracted information in a structured JSON format.
- Processes the AI response to extract and return the relevant information in the specified format.
- Purpose: Generates a structured interview plan based on the user's resume, target company, role, and interview round.
- Logic:
- Uses the Gemini model within the Agno framework to create a list of interview questions tailored to the user's selected parameters.
- Constructs a detailed prompt that guides the AI to generate questions simulating a real interview, starting with introductory questions and gradually increasing in difficulty.
- Outputs the interview plan in a structured JSON format, ensuring easy integration with other components.
- Purpose: Retrieves specific questions from an interview plan based on the serial number.
- Logic:
- Takes an interview plan and a serial number as inputs.
- Converts the serial number to a string to match the format in the JSON data.
- Iterates through the interview plan to find and return the question corresponding to the given serial number.
- Returns an error if the question is not found, ensuring robustness in handling missing data.
- Purpose: Evaluates user responses to interview questions and provides constructive feedback.
- Logic:
- Utilizes the Gemini model within the Agno framework to analyze user answers based on structure, clarity, relevance, and impact.
- Constructs a prompt that instructs the AI to evaluate answers on a scale of 1 to 10, provide feedback, suggest improvements, and decide if the user should repeat the question.
- Processes the AI response to extract and return a structured JSON format containing the score, feedback, corrected answer, and repeat status.
-
Clone the repository:
git clone https://github.com/hahaanisha/PlaceMentor
-
Navigate to the project directory:
cd Placementor -
Install the required dependencies:
pip install -r requirements.txt
-
Upload Resume: Upload your resume in PDF format. The system will parse the resume and extract relevant information.
-
Select Interview Parameters: Choose the target company, role, and interview round to generate a tailored set of interview questions.
-
Practice Interview: Use the speech recognition feature to practice answering the generated interview questions.
-
Receive Feedback: After submitting your answers, the AI agents will evaluate your responses and provide detailed feedback, including scores and suggestions for improvement.
|
|
|
|
We welcome contributions to Placementor! If you have any ideas, suggestions, or bug reports, please open an issue or submit a pull request.