First-year engineering students often struggle to connect concepts across foundational science courses such as physics, mathematics, and chemistry, leading to fragmented understanding and reduced motivation. This paper presents FIMAQUI, an interdisciplinary AI-based chatbot designed to support students during the exploratory phase of engineering education. Implemented across three university campuses, FIMAQUI integrates content from core scientific disciplines and provides 24 / 7 tutoring via natural language interaction. The chatbot was evaluated using a custom Technology Acceptance Model (TAM) survey and the validated Utrecht Work Engagement Scale (UWES-9). Results from 162 participants indicate high levels of satisfaction, clarity, and motivation, with 88.3% recommending the experience. Statistically significant improvements were observed in the Absorption dimension of academic engagement (p=0.048), suggesting deeper immersion in learning activities. These findings highlight the potential of AI tools to foster interdisciplinary thinking, enhance student engagement, and support autonomous learning in engineering education. The study contributes to the growing body of research on educational chatbots and proposes future directions for scalable, adaptive, and pedagogically integrated AI interventions.
