VisaNavig8

Inspiration

International students navigating the US visa system face an overwhelming maze of regulations, deadlines, and documents — scattered across dozens of university portals, USCIS pages, and government PDFs. A single missed deadline or misunderstood rule can jeopardize years of academic work and legal status. We built VisaNavig8 because we've lived this confusion firsthand. We wanted one intelligent, mobile-first platform that makes F-1 compliance feel less like a legal minefield and more like a guided journey.

What it does

VisaNavig8 is an AI-powered visa and immigration companion built for international students in the US. It helps students:

Navigate F-1 regulations through a searchable Wiki covering OPT, CPT, STEM OPT, I-20, RCL, travel rules, and status maintenance — sourced directly from USCIS policy manuals and university DSO offices
Track visa applications with real-time status updates and document checklists
Chat with an AI assistant trained on F-1 regulations to answer questions like "how many days of unemployment am I allowed on OPT?"
Explore destinations with visa difficulty ratings, processing times, and country-specific requirements across 180+ countries
Stay ahead of deadlines with smart alerts for application windows, travel signatures, and reporting requirements
Connect with community — share experiences and get peer advice from students who've been through the same process

How we built it

VisaNavig8 is a full-stack mobile application built with a modern, scalable architecture:

Frontend (Mobile + Web)

React Native + Expo with Expo Router for cross-platform support (iOS, Android, Web)
TypeScript throughout for type safety
Zustand for lightweight global state management
TanStack Query (React Query) for server state, caching, and 24h stale-time wiki page caching
Clerk for authentication (email/password + Google OAuth)
Custom design system: dark navy + gold theme, full dark/light mode, Apple + Linear inspired UI

Content & Data

Wiki content sourced from 39 verified URLs across Illinois Tech, RIT, and University of Chicago international student offices
USCIS Policy Manual (Volume 2, Part F) as the authoritative source
Structured WikiPage schema with tags, categories, and source attribution
Mock service layer built for instant API swap when backend is ready

Backend (in progress)

REST API: GET /wiki/, GET /wiki/search?q=
Web scraper pipeline crawling university DSO pages and USCIS chapters
AI chat layer trained on F-1 regulatory content

Challenges we ran into

USCIS content access — The USCIS policy manual bulk export URL returns 403 from the client. We had to architect around this with a server-side scraping layer and fall back to individual chapter URLs for direct linking
Regulatory complexity — F-1 rules have dozens of edge cases (cap-gap, STEM OPT employer changes, automatic revalidation). Writing accurate, non-misleading wiki content that's genuinely helpful without constituting legal advice required careful editorial judgment
Mock-to-API architecture — Designing the service layer so that swapping mock data for real API endpoints requires changing only 3 functions, with zero changes to UI components or query logic
Cross-platform consistency — Getting the dark navy + gold design system to render identically across iOS, Android, and Web with no platform-specific hacks
Search accuracy — Ensuring "unemployment" correctly returns the OPT page required a thoughtful tagging system beyond simple title matching

Accomplishments that we're proud of

Built a complete, production-quality mobile app from scratch during the hackathon — with authentication, wiki, search, markdown rendering, dark/light mode, and animated UI
Designed a clean API abstraction layer that makes the backend integration a one-day task when the API is ready
Sourced and structured real, accurate F-1 regulatory content from official USCIS and university sources — not hallucinated or generic
Created an animated hero section with pulsing globe rings and a visa stamp pop-in that captures the product's personality immediately
Achieved full dark/light mode support with a consistent design token system across every screen

What we learned

Immigration content is deeply nuanced — small wording differences (e.g. "up to 90 days" vs "exactly 90 days" of OPT unemployment) have real legal consequences for users. Accuracy matters more than speed
Designing for API-readiness from day one — even with mock data — saves enormous refactoring time later. The service layer pattern paid off immediately
Expo Router's file-based routing makes feature isolation clean and scalable; each wiki article as a dynamic route [id].tsx was the right call
Building with Clerk removed weeks of auth complexity and let us focus on the actual product problem
Good design system upfront (theme tokens, reusable components) meant every new screen took half the time of the first one

What's next for VisaNavig8

Live backend — deploying the scraper pipeline and REST API so the wiki pulls from a real, continuously updated database
AI Chat — a context-aware assistant that can answer specific questions like "I've been unemployed for 67 days on OPT, what should I do?" using RAG over our wiki content
Push notifications — deadline alerts for OPT application windows, travel signature expiry, and STEM OPT 6-month reporting
DSO directory — searchable database of Designated School Officials across SEVP-certified institutions
Document vault — encrypted storage for I-20, EAD, passport copies with expiry reminders
Multi-visa support — expanding beyond F-1 to H-1B, OPT-to-H1B transition, J-1, and green card pathways
University partnerships — direct integration with DSO portals at Illinois Tech, RIT, and University of Chicago for real-time I-20 request status