Education

University of California, Berkeley logo

University of California, Berkeley

Exchange StudentComputer Science

Aug 2025 – May 2026

Berkeley, CA

  • System related: CS 61B (Data Structures), CS 61C (Machine Structures), CS 161 (Computer Security), CS 162 (Operating Systems)
  • AI related: Data 100 (Data Science), CS 188 (Introduction to Artificial Intelligence)
National Chengchi University logo

National Chengchi University

Dual Bachelor of Arts in Risk Management & Insurance and Land EconomicsMinor in Management Information Systems

Sep 2020 – Jun 2026

Taipei, Taiwan

  • GPA: 3.86/4.00 • CS Field GPA: 3.98/4.00
  • Relevant: Data Structures (A+), Algorithms (A+), DBMS (A+), Software Engineering (A+)
Peking University logo

Peking University

Exchange Student

Feb 2024 – Jun 2024

Beijing, China

  • Relevant: Deep Learning Models and Applications, Data Visualization

Professional Experience

Ultralytics

Remote

Aug 2024 – Present

Platform

  1. Full Stack Engineer

    Jun 2026 – Present

    • Delivered dataset-wide batch annotation across the interface and model integration, including sample previews, class mapping, progress tracking, and cancellation.
    • Extended image discovery with similar-image search and embedding workflows, helping users find relevant examples across their datasets.
    • Improved dataset imports, exports, and version comparison, and contributed to model deployment workflows so teams can move from prepared data to usable models.

  2. Associate Software Engineer

    Aug 2024 – May 2026

    Part-time

    • Built annotation tools across Ultralytics HUB and Platform for detection, segmentation, classification, and oriented bounding boxes, with shortcuts, undo/redo, and reliable editing interactions.
    • Integrated SAM-assisted labeling and built the interactive embeddings explorer, connecting visual clusters and lasso selection to the image gallery.
    • Worked on shared editing and strengthened annotation workflows through end-to-end tests and release automation.

Esri R&D Center

Beijing, China

ArcGIS Earth

  1. Product Engineer Intern

    Apr 2024 – Aug 2024

    • Improved ArcGIS Earth 3D interaction by profiling rendering bottlenecks and reducing frame drops.
    • Prototyped a GPT-4o geospatial assistant to automate SQL-style queries and map generation from natural language.
    • Strengthened product stability through regression testing and code review practices for enterprise releases.

Academia Sinica

Taipei, Taiwan

History Map Project

  1. Undergraduate Researcher

    Jul 2023 – Dec 2024

    • Researched automated feature extraction for historical maps by combining YOLOv5 with Segment Anything (SAM), achieving an F1 Score of 0.90.
    • Built a full-stack vectorization & verification system (Flask, PostgreSQL, Docker, OpenLayers).

Projects

Selected work — research-driven, visual, and interactive.

Ultralytics Platform (opens in a new tab)

From raw images to deployed computer vision models

Overview

A workspace for preparing datasets, training YOLO models, and deploying computer vision applications. I contribute to annotation and image discovery, helping teams turn raw images into training-ready data.

Selected work

Contributions to annotation and image discovery.

  • Built bounding-box editing and dataset-wide annotation with model previews and class mapping.
  • Added progress tracking, cancellation, and controls for handling existing labels.
Batch annotation docs
Recorded COCO8 demo: draw a bounding box, undo and redo the edit, then save it
COCO8 · Draw, review, and save a bounding box

Historical Map of Taiwan Based on GeoAI

TGIS Best Paper
YOLOv5 + SAM (2023) · YOLOv8 Instance Segmentation (2024)

Developed an automated pipeline for digitizing historical maps using GeoAI. By advancing from YOLOv5+SAM to YOLOv8 Instance Segmentation, the system significantly improves vectorization quality, addressing the critical bottleneck of manual feature extraction in digital humanities and historical geography.

  • Labeled 8 land-use classes and synthesized training data via random crop-composition to improve generalization.
  • Upgraded from object detection to instance segmentation, improving the quality of map feature extraction.
  • Result visualizations include detection, labeling, and prediction comparisons; packaged as a PDF report.
YOLOv5 detection on historical map

YOLOv5 detection on a Japanese-era Taiwan map — grave / grass / settlement / field with confidence scores

Verification web tool

Web tool for vector verification: tile selector, detection results, and per-class counts

Ground-truth label tiles

Ground-truth tiles (red overlays mark labeled land-use)

Predicted label tiles

Model predictions on the same tiles — directly comparable to ground truth

YOLOv8 result — tree / wasteland

YOLOv8 instance results on a hatched-relief tile (tree / wasteland)

YOLOv8 result — multi-class

YOLOv8 multi-class results: field / tree / tea / bamboo / wasteland

Paper (PDF)

Pmap — The Map for Precipitation 🌦️

Crowdsourced Reports + CWA Data · Subscription & Notifications · Cloud-Native

Designed and implemented a cloud-native platform that fuses crowdsourced reports with official meteorological data to deliver real-time, high-precision precipitation mapping. The system resolves the challenge of localized weather prediction through scalable infrastructure and an interactive, real-time notification engine.

  • Fusion of public rain reports + CWA data to compute more accurate precipitation conditions.
  • Rain reporting flow with photos + real-time map experience (Leaflet) and dark mode.
  • Subscription & notification system: region-based new reports, scheduled rain updates, and threshold-based alerts (in-app + email).
Pmap end-to-end demo

End-to-end demo: report → map → subscription → notification

Real-time precipitation map

Real-time map fusing CWA station data and crowdsourced rain reports

Region-based subscription

Region subscription — pick an area + rainfall threshold + event trigger

Refining subscription region

Refining the watch region and matching event types

In-app + email notifications

Delivery: in-app banners and email when conditions match

System architecture

Architecture: VPC + Express/Redis/MySQL backend, MQTT broker, AWS Lambda for inverse-distance weather, S3 for blobs

CI pipeline

CI: PR → backend + frontend build tests in GitHub Actions → Discord status

CD pipeline

CD: build & push → registry → EC2 polls compose-prod.yml every 2 min and redeploys

GitHub Repo

Taiwan Public Bicycle System Analysis

1st Place, National Competition
Time Series + Interactive Mapping (2021) · Spatio-temporal Insights (2024)

Created an end-to-end spatial-temporal analysis framework to investigate public bicycle usage patterns. By combining time-series clustering and Random Forest feature evaluation, this study reveals critical insights into last-mile transportation dynamics and urban mobility integration.

  • Automated scraping + longitudinal station dataset (pandas) for trend analysis over time.
  • Interactive Folium map with per-station matplotlib time-series popups for exploration.
  • 2024 analysis: DBSCAN station development, last-mile + MRT transfer issues, Random Forest feature impact; won 1st Place in the National Storymaps Competition.
Space-time cube of station hot/cold spots

3D space-time cube of station hot/cold-spot intensity over time

Bike–MRT transfer flows

Transfer flows between YouBike and MRT around NTU — central stations show multi-station competition

Origin-destination flows

Origin–destination flows + trip-distance histogram

Random Forest features (1)

Random Forest features: distance to river, slope, school density

Random Forest features (2)

Random Forest features: building density, population, student count

YouBike data pipeline

Data pipeline: scrape open data → pandas → matplotlib → Folium popup

Folium popup with time-series plot

Per-station Folium popup embedding a matplotlib time-series chart

Station-density heatmap

Station-density heatmap of usage hotspots

GitHub (YouBikeAnalysis)StoryMaps (2024)Demo Website

Taiwan Convenience Store Analysis

1st Place, National Competition
Spatial Pattern (2021) · Spatio-temporal Prediction (2023)

Investigated the Hotelling's spatial competition model in urban retail networks and developed a spatio-temporal prediction pipeline. This work provides an analytical framework for identifying optimal facility locations, balancing competition and coverage in metropolitan planning.

  • ANN-based spatial pattern analysis for five major convenience store chains (Taipei metro).
  • Spatio-temporal prediction built with suitability analysis, space-time cube, and MGWR; validated statistically.
  • Won 1st Place in the National Storymaps Competition; detailed Medium write-up linked.
Suitability surface

Suitability surface combining accessibility, demographics, and competition

MGWR predicted density

MGWR-predicted convenience-store density across the metro area

Validation pipeline

Validation pipeline: suitability + OLS/MGWR + space-time cube → grid → metrics (RMSE / Moran's I / ROC-AUC)

ANN summary diagram

Average Nearest Neighbor distribution + sample point map

ANN Z-score table

ANN Z-scores by chain — dispersed individually, clustered when grouped

Collaboration vs competition density maps

Density maps: intra-chain collaboration (left) vs inter-chain competition (right)

StoryMaps (2023)GitHub (MGWR)Paper (PDF)Medium Article (2021)

Technical Skills

AI, ML & Geospatial

PyTorch
SAM
Milvus
ArcGIS
OpenLayers
R

Backend & Systems

Rust
C / C++
Python
FastAPI
Flask
Node.js
Kafka
Redis

Cloud & DevOps

AWS
GCP
Kubernetes
Docker
GitHub Actions
Jenkins
GitHub
Git

Frontend & Web

TypeScript
JavaScript
React
Next.js
Nuxt.js
Vue
Vercel
Postman

Contributions Calendar

@chuang091