Backend, full-stack, AI, and data systems.

These projects represent the main shape of my engineering work: designing APIs, building persistent AI product infrastructure, modeling relational data, and creating software that improves real research or user workflows.
For more detail, download my [resume](/Kaining-Gao-Resume.pdf) or email me at [gaokaining22@gmail.com](mailto:gaokaining22@gmail.com).

Project Focus

Kaining Gao

Python data processing pipeline and AWS-hosted FastAPI backend for automated 3D hand model analysis in a production-grade healthcare research application.

## Overview

At Iowa State University's NextGen PPT Lab, I work on software for automated 3D hand model measurement and analysis. The system supports healthcare research workflows where consistency, data accuracy, and faster measurement turnaround are central goals.

## What I Built

- Developed an automated 3D hand model analysis system in Python.
- Created a data processing pipeline that eliminated manual measurement errors and increased measurement speed by 400%.
- Maintain an AWS-hosted FastAPI backend with integrated AI processing.
- Manage PostgreSQL-backed data workflows and cloud operations for a production-grade healthcare research application.
- Collaborate with the research team to standardize measurement protocols and improve data accuracy.

## Technical Stack

- Python
- FastAPI
- PostgreSQL
- AWS
- AI processing workflows
- Data pipeline design

## Impact

This work sits at the intersection of software engineering and applied research. The most valuable part has been building tools that improve speed while protecting the quality of scientific measurement data.


3D Hand Model Analysis System

Next.js and PostgreSQL platform with persisted AI chat threads, image generation workflows, API key issuance, token metering, and OpenAI-compatible proxying for 200+ monthly active users.

## Overview

This project is a production-style AI chat and image platform built with Next.js, TypeScript, and PostgreSQL. The goal was to move beyond a simple chat demo and build the surrounding infrastructure users and developers expect: persistent conversations, account sessions, metering, quotas, API keys, documentation, and robust access control.

## What I Built

- Built persisted chat threads and image workflows in Next.js with TypeScript.
- Served 200+ monthly active users with token metering and quota-aware request handling.
- Implemented secure authentication with NextAuth sessions.
- Added developer tooling including API key issuance, usage tracking, OpenAPI documentation, and an OpenAI-compatible proxy.
- Applied production patterns including serverless API routes, provider proxying, response sanitization, client-side cache/pagination, and access control.

## Technical Stack

- Next.js
- TypeScript
- React
- PostgreSQL
- NextAuth
- Serverless API routes
- OpenAPI
- AI provider proxying

## Impact

This project gave me practical experience designing the full surface area around an AI product, not just the model call. I focused on the parts that make the system usable and maintainable: identity, quotas, auditability, persistence, and clean client/server boundaries.


AI Chat and Image Platform

Java Spring Boot backend for an event discovery application, with 20+ RESTful APIs, authentication workflows, event CRUD operations, search, and a MySQL schema designed from scratch.

## Overview

Event Finder is an event discovery application built with an Android frontend and a Java Spring Boot backend. My primary responsibility was the backend: API design, application logic, and the relational database foundation that supported user profiles, events, search, and registration data.

## What I Built

- Engineered 20+ RESTful APIs in Java Spring Boot for user authentication, event CRUD operations, search, and registration workflows.
- Designed and implemented a MySQL schema from scratch with 5 relational tables for user profiles, event details, and registration data.
- Collaborated with the Android frontend team in an Agile environment to define API contracts and keep client/server behavior consistent.
- Added validation paths for submitted event data so the backend could guard core workflows before data reached persistent storage.

## Technical Stack

- Java
- Spring Boot
- MySQL
- REST APIs
- Git
- Agile collaboration

## Impact

The project strengthened my backend engineering fundamentals: designing APIs that frontend teams can rely on, translating product requirements into relational models, and keeping service logic readable as features grow.


Projects

3D Hand Model Analysis System

AI Chat and Image Platform

Event Finder App Backend

Project Focus