I'm a second-year undergraduate student majoring in Data Science and Big Data Technology at East China Normal University.

I see myself as an AI Product Engineer in progress: someone who cares not only about models and code, but also about product experience, user needs, system reliability, and whether an AI feature actually helps people finish their tasks better.

Recently, my work has been centered around three directions:

• AI-native product engineering: AI agents, prompt engineering, tool use, product workflows, testing, and evaluation.
• Full-stack development: Java / Spring Boot, MySQL, WeChat Mini Program, Next.js, deployment, and CI/CD.
• Data & open-source analysis: graph modeling, recommendation systems, OpenDigger data, and visualization.

I learn best by building. My repositories include campus products, open-source ecosystem analysis, algorithm practice, operating-system/networking labs, and CS course projects.

Feb 2026 – Present

I'm currently working as an AIGC R&D Intern at Tezign, mainly contributing to PitchLab and Atypica.

PitchLab is an AI-powered communication practice platform for presentation training, sales conversations, interviews, and scenario-based speaking practice.

My work focuses on product-oriented AI engineering, including:

• Building and testing AI-powered interaction flows for real speaking / dialogue practice scenarios.
• Improving structured LLM outputs such as reports, feedback, analysis cards, and review results.
• Debugging product issues across AI workflow, frontend interaction, and backend behavior.
• Thinking about how AI feedback should be presented so that users can actually understand, trust, and improve from it.

Atypica is an AI research agent product that uses AI personas, interviews, and behavior analysis to help understand consumer decisions.

Through Atypica-related work, I'm learning how to build more useful AI agent systems, especially around:

• AI persona simulation and role-based interaction.
• Multi-step research workflows powered by agents.
• Human-like interview experiences and structured insight generation.
• Balancing agent autonomy, controllability, and product usability.

This internship has shaped how I think about AI products: the hard part is not only making AI respond, but making the response reliable, interpretable, and valuable in a real product context.

A campus-oriented WeChat Mini Program for team formation, group discovery, and student collaboration at ECNU.

I'm leading the product and engineering work, including:

• Designing product flows for posts, groups, discovery pages, and message systems.
• Building backend services with Java / Spring Boot / MySQL.
• Deploying and maintaining services with Linux, Nginx, GitHub Actions, systemd, and API health checks.
• Exploring AI-assisted moderation and message-pushing agents for a healthier campus community.

Goal: help students find teammates for competitions, projects, interests, and campus life more easily.

I believe good engineering is not just about writing code. It is about:

• understanding the real user problem,
• designing a clear and maintainable system,
• shipping small but reliable iterations,
• debugging patiently,
• and reflecting on what the product should become next.

For AI products, I care a lot about the bridge between model capability and user experience. A powerful model is only useful when the product can guide it, constrain it, evaluate it, and present its results in a way that users can trust.

That's why I'm especially interested in the intersection of AI agents, product design, and full-stack engineering.

• Computer networks and operating systems
• Java backend engineering and production deployment
• AI agents, tool calling, memory, and evaluation
• Full-stack product design for real student communities
• Data-driven recommendation and graph-based analysis