Potenti.Ally

Potenti.Ally: Helping Students to Connect the Dots of Who They Are

The Problem

High school students across the U.S.—especially those in under-resourced schools—are expected to make life-shaping decisions about college and careers with limited access to personal guidance. The national average student-to-counselor ratio is 376:1, with some schools lacking counseling staff entirely. This systemic gap leaves many students without structured support to explore their identities, values, and aspirations—resulting in misaligned academic and career paths, diminished motivation, and unrealized potential.

Why We Built Potenti.Ally

Potenti.Ally reimagines college and career exploration by centering the question “Who am I?” rather than just “What should I do?” It helps students reflect on their strengths, passions, goals, and learning styles through a narrative intake survey and visual identity map. These insights power a personalized, AI-guided experience that transforms self-reflection into action—offering students a sense of clarity, confidence, and purpose in planning their future.

Our goal was to democratize introspective, values-based advising—making it accessible to every student, regardless of access to traditional resources.

Technical Overview

Potenti.Ally is a full-stack web application that integrates:

• Frontend: Built with React + TypeScript and styled using Tailwind CSS for responsive, student-friendly design. Students can navigate an interactive dashboard, engage with a chatbot, and view a dynamic knowledge graph of their identity themes.
• Backend: Powered by Express.js, the server handles secure user sessions and API routing for all survey, chat, and graph operations.
• Database: A Neo4j graph database stores each student’s evolving identity map—linking strengths, passions, and goals using custom relationship types (e.g., LEADS_TO, USES, REQUIRES).
• AI Integration: OpenAI’s language model API powers a reflective chatbot that embeds identity data into every prompt, enabling highly personalized conversations. The system uses internal LLM prompts to extract and update relevant traits, skills, and themes as students continue exploring.