Inspiration
Applying to universities worldwide is complex and unequal. Students must navigate thousands of programs, deadlines, exams, and essays using fragmented tools that reset context every time. High-quality guidance is expensive and inaccessible for most students. We wanted to build a persistent AI system that compounds intelligence over time and makes professional-grade admissions support accessible globally.
What it does
AdmitPlus.AI is an AI-powered operating system for global students. It helps students plan, match, and apply to universities end-to-end using coordinated AI agents with persistent memory. The system supports university discovery and matching, application planning, and structured essay development, while continuously learning from each student’s history instead of treating every interaction as a new session .
How we built it
We built AdmitPlus.AI using Gemini models and a multi-agent architecture. Each agent has a clear role—planning, matching, evaluation, or writing—and operates over structured university data and retrieved knowledge to ensure grounded, consistent outputs. The backend is implemented in Python using cloud-native infrastructure, with a data layer that retains and reuses context across long, multi-step workflows.
Challenges we ran into
The biggest challenges were reducing hallucinations in high-stakes decisions, structuring and normalizing global university and admissions data, and orchestrating multiple agents across long-running workflows without losing consistency or context.
Accomplishments that we're proud of
We built a working multi-agent admissions system backed by a global university knowledge base and persistent student memory. The system demonstrates how AI agents can reliably automate and support complex, real-world education workflows—beyond single-prompt or session-based assistants .
What we learned
We learned that agent-based systems with clear responsibilities and grounded retrieval dramatically outperform monolithic prompts for complex, multi-step problems like education and career decision-making.
What's next for AdmitPlus.AI
Next, we plan to expand global university coverage, improve agent reasoning and evaluation, and deepen personalization for students and education partners—turning AdmitPlus into long-term learning and career intelligence infrastructure.
Log in or sign up for Devpost to join the conversation.