Interactive Career Roadmap Website: Career Paths with Progress Tracking and AI Guided Recommendations

Career navigation in rapidly evolving technology sectors presents substantial cognitive and motivational challenges for learners at all levels. Traditional static roadmaps lack adaptability to individual progress, skill gaps, and shifting industry demands. This paper presents an Interactive Career Roadmap Website — a web-based platform that integrates structured career path visualization, real-time milestone progress tracking, and AI-guided personalized recommendations using a fine-tuned large language model (LLM). The system defines four primary technology career trajectories — Data Science, Full-Stack Development, Cybersecurity, and Machine Learning Engineering — decomposed into prerequisite-linked hierarchical milestones. Users log completed milestones; the AI engine dynamically recalculates next steps by reasoning over current skill assessments, learning velocity metrics, and live industry demand signals obtained via web search augmentation. The paper details system architecture, the recommendation engine design, human-computer interaction principles applied, evaluation methodology, case studies, and open research directions for personalized career development platforms at scale.

The system architecture comprises three core components: a career path knowledge graph, a progress tracking module, and an AI recommendation engine. The knowledge graph encodes four career trajectories (Data Science, Full-Stack Development, Cybersecurity, Machine Learning Engineering) as prerequisite-linked hierarchical milestones. Each milestone includes skill assessments and learning resources. Users interact with the platform by logging completed milestones, which updates their skill profile and learning velocity metrics. The AI engine, built on a fine-tuned large language model (LLM), dynamically recalculates next steps by reasoning over the user's current skill assessments, learning velocity, and live industry demand signals obtained via web search augmentation. The recommendation engine employs retrieval-augmented generation (RAG) to incorporate up-to-date industry trends. Human-computer interaction principles guided the interface design to ensure usability and engagement. Evaluation was conducted via a randomized controlled study with 120 participants over eight weeks, comparing the interactive platform against static roadmap baselines. Metrics included goal clarity and self-reported learning motivation, analyzed using statistical tests.