Skip to content

🌁Me As A Picture

Contact

📋Résumé PDF GitHub LinkedIn ✉️ Email 📱 Phone

Work Experiences

Georgia Institute of Technology Georgia Institute of Technology Aug 2026~Dec 2026
CS 6604 Conversational AI Graduate Teaching Assistant Atlanta, Georgia
  • Deployed a Voice Agent assignment for an 80-student course: EOS→ASR→LLM→TTS pipeline with 4% ASR word error rate and EOS-to-first-audio p50 ≤ 275 ms, p95 ≤ 417 ms across four LLMs vs. 800 ms benchmark.
  • Designed RAG with cross-session memory and LoRA SFT + DPO (unit-test autograder) assignments; debugged AI eval harness so auto evals separate strong from weak submissions; held weekly office hours for student debugging.

Walmart Walmart Jun 2026~Aug 2026
Data Scientist III Intern @ Sam's Club AI Acceleration Team Bentonville, Arkansas
  • Integrated LLM & RAG with filtered vector search into a dashboard for camera/area/site-level metadata queries, guarding against hallucinations with cosine similarity & LLMs-as-a-Judge; an embedding signature verifies Milvus ingestion & retrieval use the identical embedding model, eliminating silent retrieval degradation.
  • Architected a Cassandra-sourced data layer feeding PostgreSQL and Milvus, regenerable after data loss.
  • Deployed Docker & Kubernetes CronJobs for 1,245 in-store cameras across 9 sites: monthly Gemini Vision area classification of the full fleet in 13 minutes via high-throughput concurrency, with glibc malloc arena caps and image pre-downsampling eliminating OOM failures.
  • Optimized hourly asynchronous snapshot capture to Azure, cutting 50% time and 75% storage cost via 7-day TTL.

SUEZ SUEZ May 2022~Sep 2024
Junior Data Scientist @ Smart and Environmental Solutions (SES) China Shanghai, China
  • Streamlined NLP classifier for client's imbalanced SCADA alarms, reaching 82% production accuracy.
  • Designed a DBSCAN anomaly detector for client's air compressors predictive maintenance, achieving 70% production F1 score and forecasting failures two weeks in advance despite having only a single feature.

Pramerica Life Insurance Pramerica Life Insurance Sep 2020~Apr 2022
Sep~Dec 2020: Intern; Dec 2020~Apr 2022: Data Management Trainee Shanghai, China
  • Built the company's first BI sales data platform into both PC and WeChat mini programs.
  • Developed multidimensional filtering, groupby & deduplication functions that saved 50% operation time.

Skills Bag

⛓️ LLMs & Agents
AI Infra & Platform
RAG, Hugging Face
Fine-tuning, Distillation
🪄 Gen AI & ML
Gen AI & ML
🏎️ Automation & CI/CD
Automation & CI/CD
☁️ Cloud & Data Platforms
Cloud & Data Platforms
🛢️ Databases & C++
Databases & C++
👀 Web & Visualization
Web & Visualization

Education

Georgia Institute of Technology Georgia Institute of Technology Aug 2025~Dec 2026
M.Sc. Analytics @ School of Computational Science and Engineering Atlanta, Georgia

    CategoryCourses
    Agents & LLMsConversational AI • Natural Language • Knowledge-Based AI
    DL / MLMachine Learning • ML in Computational Biology
    FrontendData & Visual Analytics • Principles-UI Software
    StatisticsMonte Carlo Methods • Nonparametric Data Analysis • Computing for Data Analysis
    BusinessBusiness Analytics Practicum • Data Analytics in Business

University of California, San Diego University of California, San Diego Sep 2019~Sep 2020
Master of Finance @ Rady School of Management La Jolla, California

National Taiwan University National Taiwan University Sep 2014~Jun 2018
Department of Economics @ College of Social Sciences Taipei, Taiwan