Tyson Zhou choutianxius

Tyson (Tianxiu) Zhou

Company	Team	Skills	Time
Meta Platforms, Inc.	Instagram Parent and Guardian Experience	AI-native Development, Python, PHP, iOS, Android, React.js	Jul 2025 - Present

Pioneer an AI-native approach to rewrite the team's in-app marketing infrastructure into a data-driven, configuration-based system, reducing code duplication by >90% and increasing new campaign setup speed by 2x.
Establish and promote an AI-first engineering cycle across scoping, planning, execution, and review, setting the standard for AI-native development within the team.
Lead the initiative to build a context management foundation for Instagram's parent-facing campaigns. Distill domain expertise into AI agent skills and plugins for automated campaign creation and testing, boosting the system's scalability.
Ship in-app marketing campaigns reaching over 100 million predicted parents on Instagram, driving measurable gains in sentiment and engagement through tight timelines with robust engineering scoping and cross-functional alignment.
Build comprehensive logging, alerting, and AI-assisted anomaly detection for Instagram's parent-facing marketing system, reducing incident response time and improving system reliability.

Project Name	Company	Skills	Time
Advanced Mobile Automated Regression Testing	Paycom LLC.	.NET, Docker, GitLab, MySQL, TypeScript, React.js	May 2024 - Aug 2024

Designed and built a .NET Core backend with MySQL for executing UI regression tests on remote iOS and Android devices, providing RESTful APIs for managing test scripts, scheduling jobs, and persisting results.
Implemented dependency injection, repository pattern, and full unit test coverage with xUnit and Moq.
Built an SPA frontend with React.js for visualizing test scripts, jobs, and results. Containerized the full stack using Docker multi-stage builds.
Set up GitLab CI pipelines to automate unit testing and linting on merge requests.
Awarded as one of the MVPs of the Software Development Internship program.

Project Name	Institute	Skills	Time
Digital Twin-Based Smart City & Smart Construction Platform	Texas A&M Univ.	TypeScript, Next.js, React Native, TailwindCSS, FastAPI, Apache Kafka, Redis, AWS	Nov 2023 - May 2025

Engineered an asynchronous data processing pipeline using Apache Kafka and Redis, with outputs persisted to AWS DynamoDB. Implemented a FastAPI-based RESTful backend to provide domain-specific real-time data analysis service.
Built responsive data visualization website and an admin dashboard using Next.js and TailwindCSS and a cross-platform mobile app with React Native.
Deployed full-stack applications using AWS services including ECS, S3, CloudFront, Route 53, and Amplify.
Established CI/CD pipelines with AWS CodeCommit, CodeBuild, and CodeDeploy to automate testing, integration, and deployment workflows.

Project Name	Institute	Skills	Time
Network Dissection of a DQN Agent for Super Mario Bros	Tsinghua Univ.	PyTorch, Deep Reinforcement Learning, OpenAI Gym, Numpy	Feb 2023 - Jun 2023

Implemented the double DQN algorithm with PyTorch to train a reinforcement learning agent to beat Super Mario Bros.
Fine-tuned a semantic segmentation model for segmenting in-game scenes to generate training dataset for the DQN agent, based on ResNet-50.
Examined the correlation between activation of neurons in the DQN network and objects in the input scenes, in an effort to investigate the mechanism of neuron function generation.

Institute	Degree	Time
Texas A&M University	Master of Computer Science	Aug 2023 - May 2025
Tsinghua University	Bachelor of Science in Biology	Aug 2019 - Jun 2023

Category	Skills
Backend Development	.NET, FastAPI, Django, Flask, Node.js, Express.js, Spring Boot, MySQL, Redis, Kafka
Frontend Development	TypeScript, JavaScript, React.js, Next.js, TailwindCSS, HTML, CSS, React Native, Redux
Professional Dev Skills	Git, Docker, CI/CD, AWS, Linux
Programming Languages	TypeScript, JavaScript, C#, Python, Java, SQL, HTML, CSS
Data Science & ML	PyTorch, Pandas, Numpy, Reinforcement Learning, D3.js, Matplotlib