I am Yan, a recent Northern Virginia Community College graduate specializing in mobile app development and software engineering, with hands-on experience in Android, iOS, Python and web technologies. Currently enhancing skills through tech fellowships with CodePath and Headstarter.

iHomie

An Android real estate mobile app. Home buyers, renters, home owners, real estate investors, and realtors can use this app to search for real estate properties, view each property's details and save favorite properties. Kotlin and Android Studio used to create this app.

FlixterPlus

An Android app that uses IMDb API and help users browse trending people today.

WordleGame

An Android app that recreates a simple version of the popular wordle game.

BitFit

An Android health metrics app that helps users to track calories of foods.

Trivia

A simple iOS mobile app that simulates trivia game. The app will calculate user’s correct answers out of the total questions at the end of the game. Swift and Xcode used to create this app.

MarvelCharacters

An Android app that that displays a random character from the Marvel API. Character Image, Character ID, description of the character show on screen when user clicks "Get the next Character" button. User can also search character by name.

TumblrDetailView

An iOS app that follows Tumblr blogs of HumansOfNewYork in table view. User can click a blog item and read content in detail view.

TipCalculator

An Android mobile app that takes user input of base amount & percentage of tips, and outputs tip amount and total bill amount. It can also take in number of persons and split bill. Kotlin and Android Studio used to create this app.

Yummy Recipes

Website of recipes. HTML, CSS used.

Inventory Tracker AI

An AI-assisted inventory management app. Users can add,delete,search for grocery items in the inventory database. Users can use webcam or mobile camera to take photos of items, and the app will automatically recognizes the items, prefill item name before user confirms addition.AI assistant can generate recipes using grocery items in inventory. Next.js, React, Firebase, Clerk, GCP,OpenAI,Vercel,CI/CD implemented.

Chat Support AI

An AI-assisted customer service chatbot for questions about Starbucks. Online Starbucks data scraped and upserted to Pinecone for RAG. Log in is optional. SignedIn users can load or clear their chat history. OpenAI, Next.js, React, Firebase, Clerk, GCP,OpenAI,Vercel,CI/CD,AWS,Caddy.

FlashcardExpo AI

Users can generate and save flashcards after signed in. They can also search, retrieve and delete previously saved cards. Stripe, OpenRouter, Next.js, React, Firebase, Clerk, GCP,Vercel,CI/CD implemented.

Find My Professor

Users can get professor recommedation with help of AI based on previously stored data. OpenAI, TypeScript, Next.js, React, JSON, PINECONE, RAG, Vercel, CI/CD implemented.

Goal Getter

AI powered task generator and goal tracker. Users can also add tasks manually. Completed tasks can be checked off and removed. Convex, OpenAI, Open Router, TypeScript, Next.js, React, JSON, Clerk,Vercel,CI/CD implemented.

Churn Prediction

Trained 8 AI models. Used 3 of them to analyze sample customer data from Kaggle. Visualize customers' probability of churning with charts. Predict whether a customer is at risk of churning, give explanation and generate email offering various incentives based on customer churning probability. Python, Streamlit, Machine Learning, Model Inference, Feature Engineering, Groq, Llama 3.2 used.

Brain Tumor Classification

Used Kaggle dataset of brain scan images to train Transfer Learning and Custom Learning models for this project. Users are able to upload a brain scan image and select a model to read the scanned image and tell if it is a tumor and what type of tumor it is. Polygraph and bar charts are used to visualize the percentage of probability and machine model confidence. Gemini AI is used to interpret the result and give user explanation.

PR Review Agent

An AI powered GitHub App that reviews pull requests to Github repos and submits reviews to users. JavaScript and Python parsers implemented. Used TypeScript, NGROK, QROQ, and Webhook.

Codebase RAG Search

Created and deployed an AI chatbot to do RAG search over two embedded codebases. Chatbot can respond to text or image query about the embedded codebases. Used Python,JupyterNotebook,Pinecone,Streamlit,sentence-transfer model,image-text parsing.