Founder & developer · built solo with Claude Code Jan 2026 — present
I’m building a personal home for fragrance: a collection, wear journal with private notes, and discovery platform. Guided taste finding, semantic search, daily picks, and researched buying options connect what someone loves with what they try next.
I designed a multi-agent system for academic and recruiting logistics. Shared intake routes events to managers and specialists; Secretary turns their verified results into a ranked digest. Explicit gates control external actions.
I built this 15-112 term project in Python with CMU Graphics, the course graphics library. Within that library’s limited capabilities, I designed the seven-class architecture, directed the Y2K Japanese game design, and debugged the collision, flipper, and launch physics. Play the browser adaptation below.
I designed the seven-class architecture, directed the game design, and debugged the physics. AI-assisted sections were disclosed under course rules, including parts of the collision math, controls, and UI.
I designed and analyzed a randomized, three-arm survey experiment with 133 participants to measure perceived value at useful-comment densities of 1, 3, or 5 of the top 10. The tested parameters informed human-review recall rules.
I developed UGC content-value standards for review and modeling teams, including categories without historical labels. I built an LLM-assisted annotation pipeline and automated recurring data collection as a scheduled agent task.
I used SQL to segment precision and recall issues by over- and under-recalled content, tracing root causes to policy gaps, human-review errors, or misleading OCR and ASR signals.
I audited video-category mappings during a search-taxonomy migration from eight levels to three tiers, flagging incompatible and overly broad mappings.
I wrote prompts and evaluated SFT and GRPO approaches for content-understanding models.
Carnegie Mellon University
Aug 2026 — present
Teaching Assistant · 15-112
I lead weekly recitations, office hours, grading, and AI-assisted programming labs for Fundamentals of Programming and Computer Science (about 15 hours/week), teaching recursion, object-oriented design, and algorithmic problem-solving.
I lead and refine an AI grading pipeline through prompt design, test-set construction, and TA review. I develop practice problems and coordinate consistent grading rubrics and student feedback.
Nutrition Alligators
Nov 2023 — Nov 2025
Co-Founder & Marketing Manager
I co-founded a Chinese nutrition company selling açaí powder and concentrate purée. I supervised 30+ factory and lab visits for product validation, audits, and quality-assurance documentation.
I shaped SKU and pricing strategy, managed and analyzed paid social and 500+ Douyin livestreams, and co-directed a 16-person team across supply chain, livestreams, and finance.
Stripe
Jul — Aug 2024 · Remote
Product & Operations Intern
I researched API gateway design, tokenization flows, and payment monetization to understand global payments infrastructure and cybersecurity.
I worked in a four-person team on competitive and benchmarking analysis of Stripe Radar, presenting our findings to Stripe executives.
03A little context
Carnegie Mellon University
B.S. in Business Administration · Additional major in Statistics and Data Science
Sophomore · Class of 2029 · QPA 4.0 Native English and Mandarin
Selected coursework
Regression Analysis · Multivariate Analysis · Probability Theory · Finance · Accounting · Programming and Computer Science
I’m interested in product management, AI, strategy, analytics, and the financial decisions that shape a business. My experience runs from product validation and pricing to model evaluation and statistical research.
Off hours
Fragrance — niche houses and white florals — F1, football, the NBA through a quant lens, fashion, and citywalks.
42 seconds · Captioned
