All work
OngoingSolo

Flyeasy

Mobile app that connects travelers with verified passengers on their exact flight.

Role
Solo founder and developer
Team
Solo
When
Apr 2026 – Ongoing
Origin
Started at the NJIT Claude Builder Club Hackathon 2026
Platforms
iOS · Android · Web
Stack
  • React Native
  • Expo
  • TypeScript
  • Supabase (Postgres)
  • AeroDataBox API
  • Claude API

Highlights

  1. 01Cross-platform (iOS, Android, web) app: save a verified flight, see who else is on it, connect and chat
  2. 02LLM feature (Claude API) that suggests places to eat, rest or spend time at your terminal, based on layover length
  3. 03Consent-first social model: connections are reciprocal and flights are verified against real airline schedules
  4. 04Privacy by design: connections are mutual (nobody messages you uninvited), a single switch hides you from all lists without deleting anything, and either side can pause a conversation without losing history
  5. 05Strict-TypeScript codebase on Supabase with a runtime-swappable theme system

Write-up

Problem

Solo travelers often share a flight with people they would happily talk to, but have no safe way to find them before boarding.

What I built

A React Native and Expo app on Supabase (Postgres) that verifies flights against airline schedules, matches passengers on the same flight, and connects them only when both sides agree. The Claude API powers a layover guide based on airport, terminal and time.

Screens

5 screens · click one to view full size

My flights
Flight detail
Find travelers
Connections
Chat

More top projects

ShippedCourse

RL for Skin Cancer Decision Support

Replicated and extended a Nature Medicine RL model for skin-lesion triage, where a missed melanoma costs far more than over-treating.

~0.96melanoma recall with the team's Dueling DQN (best epoch)

NJIT Reinforcement Learning course · Team of 3 · Replication, environment simulation, baselines, reward-matrix experiments, visualizations and presentation

  • Python
  • TensorFlow
  • PyTorch
  • Gym
  • scikit-learn