Computing students in Ugandan universities face a persistent and systemic lack of effective career guidance, which hinders their transition from academic training to meaningful participation in the technology workforce. This study introduces a novel emotion-aware conversational system designed specifically to support career guidance for university students. The proposed architecture integrates document intelligence, emotional awareness, and dynamic prompt engineering to deliver highly personalized and context-sensitive career advice. Built on a robust technical stack that includes Google Cloud services (Dialogflow CX, Vertex AI), LangChain orchestration, and the EmoRoBERTa emotion recognition model, the system provides guidance that is scalable, accurate, and responsive to users’ emotional states. Technical evaluation using BERTScore and UNIEVAL metrics demonstrates that CareerChats AI generates transparent, relevant, and actionable recommendations, effectively addressing key limitations of conventional career advisory services.
Bisaso et al. (Wed,) studied this question.