Hey there, my name is

Carey Wang

A Passionate Software Engineer

I'm a software engineer based in Irvine, CA. I thrive on building scalable backend systems, designing distributed architectures, and crafting applications that are smooth, intuitive, and impactful.

I’m passionate about creating technology that feels effortless yet powerful. From high-concurrency platforms and e-commerce systems to AI-driven orchestration tools and research applications, I love tackling challenges and delivering solutions that not only work but also inspire.

About Me Illustration

About Me

Greetings! My name is Carey Wang, and I am a passionate software engineer who enjoys turning ideas into real-world applications. Residing in Irvine, CA, I started programming during my undergraduate years in civil engineering, and since then I’ve been drawn to the endless possibilities of technology. I am currently pursuing my Master of Software Engineering at the University of California, Irvine, and will graduate in December 2025.

I began by learning the fundamentals of Java, Python, and web development. Over time, I built a strong foundation that has allowed me to quickly pick up advanced frameworks and tools such as Spring Boot, React, Redis, Docker, Kubernetes, and cloud platforms like AWS and GCP.

Along the way, I have worked on a variety of projects, including subscription commerce systems, marketing lottery platforms, AI agent orchestration engines, and transportation modeling tools. My goal is to keep honing my skills, push the boundaries of what I can build, and create technology that is reliable, scalable, and truly impactful.

Experience

Applications Programmer

Jan 2026 – Present

Institute of Transportation Studies, UC Irvine

Irvine, CA

Building AI-powered developer tooling for transportation simulation research, focusing on root cause analysis, knowledge retrieval, and production reliability.

  • • RCA Agent: Developed a post-run Python root cause analysis agent for long-running ActivitySim simulations that summarizes failure logs, extracts key stack traces and YAML config context, and generates a concise root cause report for researchers.
  • • RAG Knowledge Base: Built a retrieval-augmented knowledge base by embedding runbooks and resolved incidents into pgvector on PostgreSQL, stored source documents in Amazon S3, and returned cited snippets to suggest likely fixes.
  • • Productionization: Hardened the tool with structured log parsing, rate limiting, timeouts and fallbacks, and basic latency and token cost metrics so it runs safely on shared lab machines.

Software Engineer Intern

Jul 2025 – Sep 2025

SubMarts Inc

Irvine, CA

Built scalable backend systems for a subscription commerce platform focusing on order lifecycle management, real-time notifications, and cloud infrastructure.

  • • Designed and built an order state machine in Java to manage order transitions across payment, inventory, and notification services, avoiding broken or stuck orders by handling rollbacks under partial failures.
  • • Developed a notification service with Redis message queues and achieved a 99.9% message delivery success rate through deduplication and retry controls over 100,000 transactions.
  • • Built REST APIs using Hibernate to apply weighted scoring logic to showcase ranked customer reviews.
  • • Reconfigured AWS infrastructure with EC2 Auto Scaling Groups to handle traffic spikes.

Software Engineer (Capstone)

Sep 2025 – Dec 2025

Office of Information Technology, UC Irvine

Irvine, CA

Built a production-grade serverless interface for data lake configurations and integrated enterprise SSO authentication.

  • • Architected a serverless React and TypeScript GUI with AWS API Gateway to manage data lake configurations, replacing manual JSON workflows with a POST-based RPC API that routed validated requests to backend Lambda services backed by DynamoDB.
  • • Engineered enterprise SSO authentication by integrating AWS Cognito with UCI Shibboleth via SAML, leveraging API Gateway JWT Authorizers and custom identity claim mapping to propagate user role attributes for access control.
  • • Hardened cloud infrastructure security and efficiency by configuring API Gateway Mock Integrations for CORS preflight handling, preventing unnecessary Lambda invocations during browser OPTIONS requests.
  • • Automated the CI/CD deployment lifecycle using AWS Amplify Hosting integrated with GitHub, enabling branch-based deployments and globally distributed static asset delivery through Amazon CloudFront.

Graduate Student Researcher

Jul 2025 – Dec 2025

Institute of Transportation Studies, UC Irvine

Irvine, CA

Refactored and automated transportation simulation pipelines, improving reliability, performance, and developer experience for ongoing research.

  • • Refactored legacy transportation modeling codebase in Python, resolved dependency conflicts with Poetry dependency management and virtual environments, improved cross-platform reliability and eliminated configuration errors.
  • • Automated multi-scenario simulation pipelines with Python scripts to reduce processing time from 24 hours to 30 minutes and eliminated recurring failures.
  • • Implemented a testing and debugging setup using pytest and remote debugging, and documented troubleshooting steps that reduced system setup time by 50%.

Some Projects I've Built

ActivitySim RCA & RAG Agent

An AI-powered root cause analysis tool for failed ActivitySim transportation simulations. Uses a memory-safe reverse-chunk log parser, GPT-4o Structured Outputs, and a RAG knowledge base backed by pgvector on PostgreSQL and Amazon S3 to produce concise, cited root cause reports. Features rate limiting, exponential-backoff retries, and per-call cost metrics for production use on shared lab machines.

  • Python
  • GPT-4o
  • pgvector
  • PostgreSQL
  • Amazon S3
  • Docker
Source Code

AI-Powered Software Development Orchestrator

Developed a multi-agent AI system capable of automating the full software development lifecycle (SDLC), utilizing Model Context Protocol (MCP) and Google Gemini 2.5 models. Features an interactive Gradio web interface for iterative review and automatically generates comprehensive unit and integration tests.

  • Python
  • Gemini API
  • MCP
  • Gradio
  • Node.js
  • Docker
Source Code

PromoMall Lottery Platform

A high-concurrency marketing lottery platform built with Java, Spring Boot, Redis, RocketMQ, and ElasticSearch. Designed rule-engine microservices and implemented Redis-based sliding-block locks to ensure consistency under thousands of concurrent requests.

  • Java
  • Spring Boot
  • Redis
  • Kafka
  • ElasticSearch
Source Code

AI Agent Orchestration Platform

An orchestration system for AI agents using Spring Boot, MyBatis, Redis, and PGVector. Supports RAG knowledge bases, prompt orchestration, and hot-deploy low-code visual flows.

  • Java
  • Spring Boot
  • PGVector
  • Redis
  • Docker
Project details coming soon...

Summailize - Gmail AI Summarizer Agent

A lightweight AI agent that automatically summarizes Gmail messages into short, actionable insights. Built as a Chrome Extension with Flask backend, featuring Azure OpenAI integration, Firebase storage, and timezone-aware background scheduling for autonomous email processing.

  • Python
  • Flask
  • Azure OpenAI
  • Firebase
  • React
  • Chrome Extension
Source Code

Contact Me

If you have any inquiries, collaboration ideas, or opportunities, feel free to reach out via email. I enjoy working with passionate teams and exploring new challenges in software engineering.

Even if you just want to connect or discuss tech, I’d be happy to hear from you. Looking forward to meaningful conversations and future collaborations!

Say Hello!