Study b : A Private Local AI Practice Partner for a College Friend

StudyBuddy is a small, AI-powered practice partner developed for a college friend preparing for technical interviews, vivas, and exams. The platform was created to address the common issue of not always having someone available to practice with. StudyBuddy offers three specific modes to cater to different needs: Viva Practice, Interview Practice, and Explain a Topic.

Viva Practice mode simulates a viva examiner by presenting questions one at a time, allowing users to hone their skills in that specific area. Interview Practice focuses on technical interview topics, including Java, Data Structures and Algorithms (DSA), Database Management Systems (DBMS), and networking. Lastly, Explain a Topic simplifies complex concepts into more digestible language with relevant examples.

The user experience of StudyBuddy is streamlined through a local Streamlit application, which is easily accessible for practice. A demonstration of the application and its workflow is included with the project materials. The code for StudyBuddy is available on GitHub (https://github.com/akshat0042005/StudyBuddy).

The development of StudyBuddy relies on Python for its core programming language, Streamlit as the user interface framework, Ollama for local hosting of the AI model, and Qwen2.5 1.5B as the underlying language model for generating responses. The application employs an intentionally simple architecture, with text flow moving from the user through the Streamlit UI to the Python application, which interacts with the Ollama local API and ultimately the Qwen2.5 1.5B model, providing practice, explanation, or feedback.