MRI
MRI India Journals Vol. 13 No. 2S (2026): Special Issue: ICSAIEM

CareerPilot: Your AI-Powered Career Guidance System

Authors

  • Deepali Narwade Department of AI and DS, Dr. DY Patil College of Engineering and Innovation, Pune, India
  • Prafulla Bharate Department of AI and DS, Dr. DY Patil College of Engineering and Innovation, Pune, India
  • Om Shikare Department of AI and DS, Dr. DY Patil College of Engineering and Innovation, Pune, India
  • Ashwini Dewade Department of AI and DS, Dr. DY Patil College of Engineering and Innovation, Pune, India
  • Aishwarya Patil Department of AI and DS, Dr. DY Patil College of Engineering and Innovation, Pune, India

Keywords:

Artificial Intelligence Career Guidance System Machine Learning Skill Gap Analysis Personalized Learning Path Career Prediction LightGBM Random Forest

Abstract

Selecting an appropriate career path remains a significant challenge for students and job seekers due to limited access to personalized guidance and insufficient awareness of industry requirements. Traditional career counseling methods rely on generalized advice or academic performance metrics, which often fail to accurately represent an individual’s skills, interests, and career potential.[5] This paper presents CareerPi-lot, an artificial intelligence-based career guidance and learning platform that analyzes comprehensive user profiles to recommend suitable career paths alongside structured learning roadmaps. The system evaluates multiple attributes including skills, educa-tion, interests, experience, and resume data to predict career suitability through machine learning techniques. Additionally, the platform integrates an AI-based reasoning component using Google Gemini 2.5 Flash API that identifies skill gaps and gen-erates personalized multi-phase learning roadmaps comprising recommended resources, projects, and assessments.

The model employs a soft-voting ensemble combining LightGBM and Random Forest classifiers trained on 2,100 synthetic user profiles distributed across 30 career categories. Features are extracted using TF-IDF text embeddings (300 dimensional skill features, 50-dimensional industry features) combined with one-hot encoded categorical variables and numerical indicators, resulting in a 365-dimensional feature space. The platform employs React and TypeScript for the frontend interface and Django with Django REST Framework for backend services. Experimental evaluation on unseen test data demonstrates 90% overall accuracy, 99% top-3 accuracy, and 0.93 weighted F1-score, validating the effectiveness of the proposed approach in bridging the gap between academic learning and industry demands.

 

Downloads

Published

2026-06-15

How to Cite

Narwade, D., Bharate, P., Shikare, O., Dewade, A., & Patil, A. (2026). CareerPilot: Your AI-Powered Career Guidance System. Multidisciplinary Journal of Research in Engineering and Technology, 13(2S), 54–60. Retrieved from https://journals.mriindia.com/index.php/mjret/article/view/3553

Similar Articles

<< < 1 2 3 4 5 6 7 8 9 10 > >> 

You may also start an advanced similarity search for this article.