Skip to main content Skip to navigation

Sakshm AI: Advancing AI-Assisted Coding Education for Engineering Students in India Through Socratic Tutoring and Comprehensive Feedback

Project Overview

The document examines the use of Sakshm AI, an innovative AI-assisted coding education platform designed for engineering students in India, emphasizing the transformative potential of Large Language Models (LLMs) in educational settings, particularly in programming. By leveraging a Socratic tutoring approach, Sakshm AI facilitates personalized feedback and adaptive learning, enhancing critical thinking and independent problem-solving skills among learners. Key features of the platform include the Disha chatbot, which supports students in navigating coding challenges. Research findings reveal that while Sakshm AI significantly improves engagement and learning outcomes, it also faces challenges related to user engagement and technical performance. Overall, the document underscores the promising role of generative AI in revolutionizing education by fostering a more interactive and personalized learning experience, despite some existing hurdles.

Key Applications

Sakshm AI - Intelligent Tutoring System

Context: Coding education for engineering students in India, particularly for undergraduate learners.

Implementation: Implemented as a web-based platform that integrates AI-driven components, including a chatbot for Socratic tutoring and feedback.

Outcomes: Improved problem-solving skills, enhanced engagement, personalized learning experiences, and support for diverse learners.

Challenges: Technical issues, limited question variety, and the potential for decreased independent thinking due to reliance on AI assistance.

Implementation Barriers

Technical Barrier

Users reported technical issues such as slow performance and bugs that disrupt the learning experience.

Proposed Solutions: Implementing an autosave feature, enhancing system stability, and ensuring fast response times.

Content Barrier

Limited variety in coding problems and lack of real-world application scenarios were noted by users.

Proposed Solutions: Expanding the question bank to include more diverse and complex problems, and incorporating real-world scenarios.

Project Team

Raj Gupta

Researcher

Harshita Goyal

Researcher

Dhruv Kumar

Researcher

Apurv Mehra

Researcher

Sanchit Sharma

Researcher

Kashish Mittal

Researcher

Jagat Sesh Challa

Researcher

Contact Information

For information about the paper, please contact the authors.

Authors: Raj Gupta, Harshita Goyal, Dhruv Kumar, Apurv Mehra, Sanchit Sharma, Kashish Mittal, Jagat Sesh Challa

Source Publication: View Original PaperLink opens in a new window

Project Contact: Dr. Jianhua Yang

LLM Model Version: gpt-4o-mini-2024-07-18

Analysis Provider: Openai

Let us know you agree to cookies