Wuhan University
—
My master's work focuses on knowledge retrieval and agent workflows. I am also exploring how to combine long-term memory with RSI so that systems can retain and use valuable information and improve iteratively based on feedback.
Me
About Me
You can call me Julian or Jules, both loosely adapted from Jue, the pinyin spelling of my name. Online, I've long gone by hyacehila. It comes from hyacinth, a flower I like, with endings like -hila / -ila that sound a little like a sprite's name to me. I sometimes picture it as a sprite flying through a sea of hyacinths, curious about whatever it finds along the way.
That's also close to how I think about technology and engineering. A new technology may be capable of a lot, but where it belongs and whether it helps are things we have to find out in practice. When building agents, I start with the business and how people currently do the work. I look at what needs model judgment and what workflows, retrieval, or tools can handle. Then I make the idea and expectations clear, build something, see where the results differ, and keep adjusting. Models need some room to explore, but we need ways to check the results and learn from what goes wrong.
Lately, I've been interested in knowledge systems, AI4Game, long-term memory, and AI4Fin. When work continues over time, can AI remember what's already been done, make sense of human feedback, and use that experience in its next decisions? I want to build systems that can work with people in that way over time. I use this blog to record what I read, what I try, and what I haven't figured out yet. There's fiction here too. I enjoy imagining characters, wondering what choices they'll make, and letting the story unfold.
Education
Xidian University
—
My undergraduate studies in mathematics, statistics, and programming helped me develop ways to analyze problems. I enjoy putting what I learn into practice and, when results differ from expectations, examining the process and revisiting my reasoning.
Work and Internship
AI Agent R&D Engineer (Intern)
NetEase Interactive Entertainment (Shanghai) · June–August 2026 · Internship retrospective
I worked on AI agent development for games, including UI engineering automation and the organization and retrieval of game knowledge. I was interested in how agents could reduce repetitive work for developers and whether their outputs could be used in real projects.
Algorithm Researcher (Intern)
NSFOCUS Technology (Wuhan) · December 2025–March 2026
I researched agents for code security, including vulnerability analysis, tool-assisted verification, and post-training data preparation. I focused on the verifiability of model judgments and on whether agents could adjust their analysis in response to tool feedback.
Research
- Unveiling the Drivers of PTSD: An Interpretable Machine Learning Approach with SHAP International Conference on Intelligent Computing and Data Analysis 2025 ; EI DOI
Awards and Certificates
- First Prize, Shaanxi Province — National College Student Statistical Modeling Competition
- Third Prize, National Level — SAS China University Data Analysis Competition
- Second Prize — Mathematical Contest in Modeling / Interdisciplinary Contest in Modeling (MCM/ICM)
- College English Test: CET-4 510 | CET-6 513