diff --git a/_data/cv.yml b/_data/cv.yml
index 07167c3bef95b..73425a00bd8ac 100644
--- a/_data/cv.yml
+++ b/_data/cv.yml
@@ -4,11 +4,11 @@
- name: "Name:"
value: Gary Goh Shing Wee
- name: "Experience:"
- value: "3 years academic experience in machine learning and predictive modeling + 1 year industrial experience in building a complex recommendation system for an online video live-streaming and community platform"
+ value: "8+ years across GenAI, machine learning, and data platform engineering — spanning the full data function: data engineering, analytics, data science, and MLOps deployment"
- name: "Interest:"
- value: Data-driven Artificial Intelligence applications
+ value: Data-driven and agentic AI applications; data & AI governance
- name: "Proficiency:"
- value: Large-scale data processing using Hadoop and Spark SQL and training deep neural networks with deep learning frameworks such as TensorFlow and PyTorch
+ value: Architecting governed data warehouses on Databricks with CDC streaming via AWS DMS, and shipping LLM / agentic systems to production (built with TensorFlow and PyTorch)
- name: " :"
value: garygsw@gmail.com
- name: " :"
@@ -17,6 +17,19 @@
- title: Work Experience
type: table
contents:
+ - title: Data Scientist
+ subtitle: "WhiteCoat — SEA digital health platform (SG/MY/PH/HK/ID)"
+ year: 2024-Present
+ start_date: May 2024
+ end_date: Present
+ location: Singapore
+ description:
+ - Work across the full data function end to end — data engineering, analytics, data science, and AI transformation — supporting management decision-making and fundraising
+ - Architected Data Warehouse 2.0 on Databricks, migrating to a governed native catalog with CDC streaming via AWS DMS and cutting data-platform costs by ~80%
+ - Designed a company-wide AI operating model — a git-based knowledge base plus a governed MCP layer connecting AI agents to live company data (Databricks, Jira, GitHub, Drive, Gmail, Mixpanel) — and rolled out Claude and GPT/Codex org-wide with read-only and PII-ringfencing guardrails
+ - Built a self-healing LLM pipeline that auto-classifies Databricks job failures, self-repairs or opens PRs, and closes Jira tickets, eliminating manual triage
+ - Deployed predictive models, including quantile-regression doctor-demand forecasting, an AI Fraud/Waste/Abuse detection engine across 14 risk categories, and B2B corporate segmentation
+ - Data lead for the Series C fundraise data room; delivered self-serve natural-language analytics via Databricks Genie and an MCP-connected agent
- title: Data Scientist
subtitle: "Sea Ltd — Garena"
year: 2021-2022
@@ -97,20 +110,24 @@
contents:
- name: " GitHub:"
value: "https://github.com/garygsw"
+ - name: "GenAI / ML:"
+ value: LangChain, LangGraph, MCP, RAG, agents, Claude Code, TensorFlow, PyTorch, scikit-learn, MLflow
+ - name: "Data & Platform:"
+ value: Databricks, Apache Spark/Hive/Hadoop, AWS DMS (CDC), Unity Catalog, ETL pipeline design
+ - name: "Governance:"
+ value: AI & data governance, Unity Catalog access control, PII ringfencing, read-only guardrails
- name: "Programming languages:"
- value: Python, C/C++/C#, Javascript, Java, Microsoft VBA, R
- - name: "Operating systems:"
- value: Mac OS, Windows, Ubuntu, Raspberry Pi OS, Linux Mint
+ value: Python, SQL, Bash, Javascript, C/C++/C#, R, VBA
+ - name: "Cloud:"
+ value: AWS (S3, EC2, Lambda, SageMaker, Bedrock, AgentCore), GCP (GCS, Vertex AI)
- name: "Python libraries:"
- value: PyTorch, NumPy, Pandas, SciPy, Keras, Theano, scikit-learn, SpaCy, Django, ...
- - name: "Frameworks:"
- value: Jupyter, Qt, AWS, Miscrosoft .NET 4.5, Cordova
+ value: PyTorch, NumPy, Pandas, SciPy, Keras, scikit-learn, SpaCy, Django, ...
- name: "Database systems:"
- value: MySQL, PostgreSQL, SQLite, MongoDB
- - name: "Client-side technologies:"
- value: HTML5, CSS3, JQuery, D3.js, AngularJS
- - name: "Other software:"
- value: Automod, AIMMS, Adobe Photoshop and Illustrator
+ value: MySQL, PostgreSQL, MSSQL, MongoDB, Databricks
+ - name: "Visualization:"
+ value: Grafana, Superset, Metabase, Lakeview, Power BI, Tableau, Matplotlib, D3.js
+ - name: "Operating systems:"
+ value: Mac OS, Windows, Ubuntu, Raspberry Pi OS, Linux Mint
- title: Others
type: list
diff --git a/_news/announce_whitecoat.md b/_news/announce_whitecoat.md
new file mode 100644
index 0000000000000..8db626d549543
--- /dev/null
+++ b/_news/announce_whitecoat.md
@@ -0,0 +1,7 @@
+---
+layout: post
+date: 2024-05-01 00:00:00+0800
+inline: true
+---
+
+I joined [WhiteCoat](https://www.whitecoat.global) as a Data Scientist, working across the full data function — from data-platform engineering to company-wide AI transformation!
diff --git a/_pages/about.md b/_pages/about.md
index b5b00cd08f758..a6d0e06731bae 100644
--- a/_pages/about.md
+++ b/_pages/about.md
@@ -3,15 +3,14 @@ layout: about
title: about
nav-title: about
permalink: /
-description: Data Scientist • Garena • Sea
+description: Data Scientist • WhiteCoat
profile:
align: right
image: prof_pic.jpg
address: >
1 Fusionopolis Place #17-10
-Singapore 138522
+Singapore
news: true # includes a list of news items @@ -19,7 +18,9 @@ selected_papers: true # includes a list of papers marked as "selected={true}" social: true # includes social icons at the bottom of the page --- -Hello 👋 Welcome! I am a Data Scientist at [Garena](https://www.garena.sg), and I love to explore data-driven approaches to solve real-world problems. +Hello 👋 Welcome! I am a Data Scientist at [WhiteCoat](https://www.whitecoat.global), a Southeast Asian digital health platform (SG/MY/PH/HK/ID), where I work across the full data function end to end — data engineering, analytics, data science, and company-wide AI transformation. I love using data-driven and agentic-AI approaches to solve real-world problems. + +Before WhiteCoat, I was a Data Scientist at [Garena](https://www.garena.sg) (Sea Ltd), building recommendation systems for an online video live-streaming and community platform. I got my M.Eng. from [Singapore University of Design and Technology](https://www.sutd.edu.sg) [ISTD Pillar](https://istd.sutd.edu.sg), advised by [Alexander Binder](https://www.visual-intelligence.no/people/alexander-binder) and [Kwan Hui Lim](https://sites.google.com/site/limkwanhui/). I did research on explainable artificial intelligence (xAI) methods for interpreting deep learning models. diff --git a/assets/pdf/CV.pdf b/assets/pdf/CV.pdf index 8e465bac98f30..c9f51d2a0ebcd 100644 Binary files a/assets/pdf/CV.pdf and b/assets/pdf/CV.pdf differ