Throughout the hackathon, one of the major challenges we faced was managing time while building a full end-to-end system that included agentic AI logic, real-time interaction, and a clean user experience. Designing the agentic interview flow using Kiro required multiple iterations to balance realism, adaptability, and system stability. Another hurdle was integrating real-time transcript generation with interview flow control, ensuring that user responses were accurately captured without interrupting the interview experience. Additionally, coordinating cloud services, handling session state correctly, and debugging edge cases under tight deadlines demanded rapid decision-making and trade-offs. Despite these challenges, the constraints pushed us to prioritize core features, improve system design clarity, and deliver a scalable and functional solution within the limited hackathon timeframe.