Year: 2025
Type: article
Abstract: Traditional career counseling methods often fail to keep pace with evolving job markets, leading to misaligned guidance and decision fatigue. This paper introduces a comprehensive, AI-powered system that integrates a multi-agent architecture with locally hosted Large Language Models (LLMs) via Ollama to deliver personalized, privacy-preserving career planning. Key system components include: • Profile Agent: Structures user inputs such as academic background, interests, and constraints into a machine-readable profile. • Career Agent: Generates context-specific career suggestions with justification, demand forecasting, and comparative pros/cons. • Skills Agent: Maps chosen career paths to prioritized skill sets, certifications, and curated learning resources. • Roadmap Agent: Synthesizes actionable, timeline-based roadmaps tailored to individual goals and resource constraints. Built with Streamlit for an interactive web interface, the system was evaluated with 20 users across diverse backgrounds. Results show 85% high relevance in career suggestions, 90% alignment in skill mapping, and an average end-to-end latency under 20 seconds. The prototype demonstrates significant gains in adaptability, explainability, and user satisfaction compared to traditional tools. Future work includes multilingual support, dynamic labor-market integration, and gamified progress tracking to further enhance user engagement and real-world applicability.
Authors: Sanat Ladkat, Manisha Bharati
Institution: Savitribai Phule Pune University
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: hybrid
OpenAlex ID: W4410928760