Preparing students for the workforce requires aligning their skills and interests with career opportunities. This research presents an AI-driven system using machine learning and multimodal large language models to support workforce development. The system automates resume classification and provides personalized feedback, including course recommendations, skills gap analysis, learning pathways, and training suggestions. It identifies best-fit roles for students and offers actionable insights to boost career readiness. The system consists of two main phases: resume classification and workforce analysis. After using machine learning for resume classification to identify the best-fit role, the system utilizes the Generative Pre-trained Transformer (GPT-4) to provide contextual feedback on skills gaps, recommend learning pathways, and identify training opportunities. Two different GPT-4 instances are used simultaneously in the system. One with general knowledge of various fields and information on the required skills, training, and career development. The other GPT-4 instance has domain-specific knowledge of specific university programs, courses, internal policies, and procedures. The domain-specific GPT-4 knowledge base is a pluggable component that can easily integrate and change from one institution to another, allowing tailored, customized support for each school and its students. To evaluate the automated system, we used a dataset consisting of 962 resumes that were manually labeled. We extracted key text-based features from each resume using natural language processing techniques, like term frequency-inverse document frequency (TF-IDF) and word to vector (Word2Vec). Then, we trained and tested various machine-learning models for resume classification, including Logistic Regression, Random Forest, Support Vector Machine (SVM), K-nearest neighbors (KNN), and an Ensemble model. The test results indicated that the Random Forest was the most robust model, and the system achieved an accuracy of 99.4% in resume classification and identifying the best-fit roles. The inspection of the GPT-4 responses indicated high accuracy, robustness, and consistency in providing expert-level feedback. The implications of the research work and the implemented system extend to multiple stakeholders. It benefits students through precise career guidance. Additionally, academic advisors can use it to identify learning pathways and recommend training programs, while HR professionals can assess candidate readiness and streamline hiring decisions.