Issuu

International Research Journal of Engineering and Technology (IRJET)

e-ISSN: 2395-0056

Volume: 12 Issue: 12 | Dec 2025

p-ISSN: 2395-0072

www.irjet.net

Beyond Autocomplete: A Multimodal RAG-Enhanced AI Tutoring System for Validated Programming Education S D Keerthiga Devi1, Nivedita Raushan2, Palak Pamnani3, Jyoti Kumari4 1,2,3,3 Student, KCC Institute of Technology, Greater Noida, Uttar Pradesh, India

---------------------------------------------------------------------***---------------------------------------------------------------------

Abstract - Recent large-scale surveys reveal a disturbing

generated by AI is secure, while 17.7% of them identified output inaccuracy as their top concern [1]. These results point to a core disruption in the traditional skill-building pipeline [10] [21].

paradox in programming education: whereas 84.2% of developers now use AI coding assistants, these tools produce incorrect code in 40% of complex situations. Even worse, novice programmers detect only 27% of such mistakes, embedding misconceptions that last for more than 18 months and fundamentally undermine their skill development. This paper introduces Code Mentor, a multimodal AI tutoring system that revolutionizes the way students learn to program while preserving educational integrity. This system offers key innovations: the Semantic Code Similarity Detection Algorithm (SCSDA); an unprecedented four-stage validation framework that checks not only the accuracy but also the educational value of code suggestions, expected to achieve accuracy of over 90% due to strict pedagogical validation; Validated AI Learning Theory (VALT), the first ever holistic theoretical framework designed for AI-enhanced programming education that synthesizes key established learning theories to drive intelligent instruction; full VARK learning model integration and two pioneering metrics-the Learning Velocity Index (LVI) and the AI Dependency Coefficient (ADC). Code Mentor will achieve high educational impact due to effect sizes greater than Cohen's d > 1.0, fundamentally bridging the gap between convenient AI support and authentic programming mastery.)

1.2 Research Problem Current AI coding assistants are optimized for productivity [13], rather than pedagogical effectiveness [2, 18], leading to five critical shortcomings: 1.

Key Words: AI programming education, code validation, educational technology, learning analytics, multimodal learning, RAG systems

1. INTRODUCTION

1.3 Our Proposed Solution

1.1 The AI Education Crisis

The design and architecture of Code Mentor, a comprehensive AI tutoring system created to fill these crucial gaps using the following innovations, are described in this paper:

The fast development of AI in software development poses some serious challenges for programming education [2] [18]. Surveys with extensive responses from 481 professional programmers show that 84.2% of them use AI coding assistants like Git Hub Copilot (37.9%) and Chat GPT (72.1%) [1]. While these tools have been widely adopted [9] [15], many educational issues remain unresolved [2] [10].

Empirical evidence underlines the severity of the problem: 40% of AI-generated code is erroneous in complex scenarios; novice programmers spot only 27% of those errors [9]. Only 23% of developers believe the code

Problem 1: Validation Gap: Tools generate code that is unverified [16], thus, making novice learners maintain misconceptions that prevail in 67.8% of the errors [12]. Problem 2: Context Blindness - 14.4% of developers refer to "lack of context understanding" as a barrier [1], which AI is unable to integrate course materials or learning objectives [4] [10]. Problem 3: Single Modality - Current tools are textbased [6] [14] despite evidence that 36% of learners prefer visual, 26% prefer auditory, and 15% prefer kinesthetic approaches. Problem 4: Analytics Absence - Currently, there is no comprehensive tool for tracking student progress, locating weak learners, or monitoring AI dependency [11] [15]. Problem 5: Missing Pedagogical Theory - The systems developed so far are not supported by any theoretical learning science or principles of cognitive development [2] [18].

Impact Factor value: 8.315

New Framework for Validation: The Semantic Code Similarity Detection Algorithm (SCSDA) is a multiphase procedure that verifies code recommendations for accuracy and alignment with instructional objectives [16] [22]. Full Multimodal Integration: The first thorough application of the VARK model in programming education [2].

ISO 9001:2008 Certified Journal

Page 949