StudyPort helps students navigate one of the most consequential decisions of their academic life: which university abroad to apply to, and how to actually perform in the interviews that decide admission. We built the AI layer behind both halves of that journey - a chatbot that recommends universities based on a student's profile, and a video interview tool that assesses and coaches students before the real thing.
Technology Stack
The Challenge
Choosing a university abroad involves matching academic background, budget, target country, and career goals against thousands of possible programs - more combinations than a static filter or a generic counselor conversation can meaningfully cover. On the admissions side, video interviews are often a student's first experience with that specific format, with no low-stakes way to practice and get specific feedback before the interview that actually counts.
What We Built
- University recommendation chatbot: a conversational AI that asks about academic history, budget, target countries, and career goals, then recommends a ranked shortlist of universities and programs
- Conversational refinement: students can question or push back on recommendations ("cheaper options," "closer to a city") and the chatbot re-ranks in real time instead of restarting the flow
- AI video interview simulator: presents students with realistic admissions-style interview questions and records their video responses
- Automated response assessment: analyzes speech content, clarity, and delivery from the recorded video to score answers and flag specific weak points
- Interview readiness coaching: generates targeted feedback and suggested improvements per question, so students practice against their actual weak points instead of generic tips
How It Works
The recommendation chatbot uses natural language processing to extract structured criteria - budget, country preference, academic scores, career direction - from an ordinary conversation, then ranks universities against that profile and explains why each recommendation fits. For interview practice, computer vision and NLP models process the student's recorded video response together, evaluating both what was said and how it was delivered, and the platform turns that analysis into specific, actionable coaching rather than a single pass/fail score.
The Outcome
Students get a university shortlist grounded in their actual profile instead of generic rankings, and a way to rehearse admissions interviews with concrete feedback before the interview that matters. For StudyPort, it turns two of the most anxiety-inducing steps in the study-abroad process - choosing where to apply and preparing for the interview - into a guided, AI-assisted experience rather than a black box.