Scribd
Upload
10 views2 pages

AI Mock Interview Evaluator Plan

The AI Mock Interview Evaluator Project aims to create a secure, open-source platform for mock interviews. Key steps include finalizing API validation, integrating fluency metrics, enhancing qualitative feedback, implementing a basic database for session storage, and creating a history API. The project also involves setting up the interview flow, capturing audio, securely transmitting data, and displaying comprehensive results to users.

Uploaded by

xijewa3874
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
10 views2 pages

AI Mock Interview Evaluator Plan

The AI Mock Interview Evaluator Project aims to create a secure, open-source platform for mock interviews. Key steps include finalizing API validation, integrating fluency metrics, enhancing qualitative feedback, implementing a basic database for session storage, and creating a history API. The project also involves setting up the interview flow, capturing audio, securely transmitting data, and displaying comprehensive results to users.

Uploaded by

xijewa3874
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

🏗️ AI Mock Interview Evaluator Project Plan

Project Goal: Build a secure, comprehensive AI Mock Interview


Evaluator (Open-Source Edition).

Step Focus Area Action Details

B1. Validation & Finalize API Validation: Ensure every secure endpoint in
Logic your running local_api.py checks for the custom
LOCAL_API_KEY (sent via the X-API-KEY header)
before processing data.

B2. Fluency Combine Metrics: Modify the /api/evaluate endpoint to


Integration correctly accept the required time_taken (in seconds)
from the client and integrate the analyze_fluency
function. This allows the API to return a combined report
(Content + Fluency).

B3. Refine Enhance Qualitative Advice: Update your


Feedback analyze_fluency function to provide text-based,
action-oriented advice (e.g., "Your WPM is 80, which is
too slow. Try to be more concise.") based on the WPM
and filler counts.

B4. Data Implement Basic Database: Integrate a simple database


Persistence (like SQLite) and create a table to store the full JSON
report for every interview session (Session ID, Role, Date,
and the JSON Report).

B5. History Create History API: Add a new GET endpoint (e.g.,
Endpoint /api/history) to query the database and fetch a summary
list of past interview sessions.
Step Focus Area Action Details

F1. Interview Flow Create client_app.py: Write the main user script
Setup that sequences the interview: Prompts user for Role
→ Calls API to get Question (e.g., /api/question) →
Displays question.

F2. Audio & Time Finalize Microphone Input: Use


Capture SpeechRecognition to accurately capture audio and
record the exact time duration of the answer. This
script is responsible for providing the time_taken
value.

F3. Secure Send Data to API: Use the requests library to


Transmission package the transcription, question, and time
duration, and POST this data to the backend's
/api/evaluate endpoint, including the
LOCAL_API_KEY in the header.

F4. Comprehensive Parse & Display Results: Receive the final


Reporting combined JSON report from the API. Parse all the
scores (Content, STAR, WPM, Filler Count) and
format them into a clear, professional, easily
readable final report for the user.

F5. History Retrieval Show Past Sessions: Add a menu option to


client_app.py to call the backend's /api/history
endpoint and display a simple list of the user's past
interview sessions.

You might also like