Loh Rui Jie Joel

Bridging data science, AI, and modern web design — with a touch of elegance.

Available for projects
Get To Know Me

About Me

Crafting digital experiences with passion, precision, and purpose

👋Hello, I'm Joel Loh

A passionate Data Scientist and  Gen AI Web Developer who loves building things — from AI-powered tools to full-stack apps that make data more accessible and experiences more intuitive.

My work blends analytics, AI, and full-stack development — from building generative AI tools and dashboards to deploying React apps with FastAPI and Docker. I care about clarity in both code and user experience.

Whether I'm experimenting with LangChain, integrating vector databases, or debugging edge-case UI issues, I try to keep things grounded and purposeful. I'm always learning, and always building.

Core Technologies

DatabricksMongoDBAstra DBPostgreSQLOpen AI APIStreamlitTableauUbuntu ServerRenderVercel

Business Analyst Intern

Health Promotion Board

Jan 2025 - May 2025

Led data-driven improvements for the Healthy 365 app by analysing user behavior patterns and training machine learning models to optimise personalised activity goals.

Collaborated across product and analytics teams to turn insights into impactful recommendations for public health engagement.

ML • Data Analysis • Pyspark

Full Stack Web Developer

Project SAVE

Jan 2024 — May 2024

Designed and implemented a full-stack conversational platform that leveraged large language models (LLMs), integrated with a React + Tailwind frontend.

Handled deployment, authentication, and backend routing with MongoDB database to enable context-aware, real-time feedback for church outreach engagements.

Next.js • Docker • LLMs

Let's Connect

joellohrj@gmail.com
jlorj.dev
Singapore

Available for Hire

Open to New Opportunities
Open to Relocation
Remote & On-site Available
Tech Arsenal

Technology Stack

Technologies I use to bring ideas to life

AI & Machine Learning

Python

Expert
4 years exp

Scikit-learn

Intermediate
2 years exp

Pytorch

Intermediate
2 years exp

Classification

Intermediate
2 years exp

Regression

Intermediate
2 years exp

Clustering Algorithms

Intermediate
2 years exp

Statistics & Analytics

Pandas

Expert
4 years exp

SQL

Intermediate
3 years exp

MS Excel

Intermediate
3 years exp

R

Intermediate
2 years exp

PySpark

Intermediate
2 years exp

Tableau

Beginner
1 year exp

Frontend

React

Intermediate
2 years exp

Next.js

Intermediate
2 years exp

Tailwind CSS

Intermediate
2 years exp

JavaScript

Intermediate
2 years exp

TypeScript

Beginner
1 year exp

Backend

FastAPI

Beginner
1 year exp

Streamlit

Beginner
1 year exp

Node.js

Beginner
1 year exp

Express

Beginner
1 year exp

Database

MongoDB

Intermediate
2 years exp

AstraDB

Beginner
1 year exp

PostgreSQL

Beginner
1 year exp

Cloud & DevOps

Git

Expert
4 years exp

Docker

Intermediate
2 years exp

Caddy

Beginner
1 year exp

Vercel

Beginner
1 year exp

Render

Beginner
1 year exp
Professional Journey

Experience

My professional journey and the impact I've made across different domains

Jan 2025 — May 2025

Business Analyst Intern

Health Promotion Board (HPB)

Worked cross-functionally to monitor and improve user engagement and segmentation accuracy within real-time analytics pipelines.

Key Achievements

Developed anomaly detection systems to monitor model drift and segmentation inconsistencies
Collaborated with developers and health officers on product decisions backed by analytics
Adapted to evolving analytics requests in an ambiguous data environment

Technologies & Skills

SQLPandasStatistical AnalysisData VisualisationPython
Jun 2023 — Dec 2023

Data Scientist Intern

Health Promotion Board (HPB)

Designed machine learning pipelines and clustering models to evaluate the effectiveness of lifestyle programs in the Healthy365 app.

Key Achievements

Built segmentation models using HDBSCAN, DBSCAN, and K-Means to identify behavior-based user clusters
Engineered scalable data pipelines in PySpark for predictive analytics
Delivered 5 key insights that shaped app feature enhancements
Presented findings to cross-functional teams to influence program strategy

Technologies & Skills

PythonScikit-learnPySparkPandasDatabricksData Analytics
Portfolio Showcase

Featured Projects

Discover my latest work in web development, AI integration, and digital innovation

E-Commerce
Completed2 weeks

Kpop Franchise E-Commerce

Modern Kpop Merchandise Storefront

A full-stack e-commerce platform for a Kpop merchandise brand. Designed with a clean, minimal aesthetic and a robust architecture for product listing, cart, and checkout flows.

View RepositoryLive Demo

Key Features

  • Implemented product catalog, cart management, and order summary flows using Stripe API
  • Built responsive, mobile-first UI with Tailwind CSS and Shadcn UI components
  • Integrated Supbase for storing products, users, and order data
  • Set up secure user authentication and protected routes with NextAuth
  • Optimized server-side rendering (SSR) with Next.js for fast page loads

Deliverables

  • Complete multi-page e-commerce site with shop, product details, and cart pages
  • Reusable product and checkout components for future scalability
  • Supabase data schema for products and user sessions
  • Deployed on Vercel for production environment

Technology Stack

Next.jsReactSupabaseTailwind CSSStripeNextAuthShadcn UI
Social Impact
Completed4 months

Project SAVE

Conversational Outreach AI Platform

A full-stack sentiment analysis platform designed to deliver real-time AI-powered insights from outreach conversations. Built with an emphasis on usability and deployment under constrained environments.

View RepositoryLive Demo

Key Features

  • Trained and evaluated 5 sentiment classifiers (F1 ↑ to 80%)
  • Automated web scraping pipeline for training data
  • Integrated MongoDB for persistent storage
  • Dockerised deployment with Node.js and Caddy
  • CI-ready Git workflow for collaborative development

Deliverables

  • Fully deployed MVP with web interface and real-time predictions
  • Next.js frontend and Node.js backend integration
  • MongoDB schema and pipeline optimisation
  • Production-ready Docker + Caddy setup

Technology Stack

Next.jsReactMongoDBDockerNode.jsCaddyPython
AI Chatbot
Completed2 weeks

Budget 2024 AI Chatbot

Government Budget Q&A Assistant

An AI-powered chatbot designed to respond to questions about Singapore’s Budget 2024 in real time. Built using LangChain.js and GPT models to provide accurate, context-aware answers.

View RepositoryLive Demo

Key Features

  • Deployed on Vercel with serverless infrastructure
  • Integrated OpenAI GPT-4 with LangChain.js for retrieval-augmented generation
  • Real-time query handling with sub-2 second response time
  • Context-aware semantic search using Astra DB
  • Conversational memory and document parsing capabilities

Deliverables

  • Frontend UI with Next.js and streaming output interface
  • LangChain-powered document loader and retriever
  • Production deployment via GitHub Actions
  • Public open-source repository hosted on GitHub

Technology Stack

Next.jsLangChain.jsOpenAI GPT-4Astra DBVercel
LLM Infrastructure
Completed1 week

EV AI Chatbot Assistant

GenAI Web Agent with FastAPI + LangChain

Developed a lightweight Generative AI backend for electric vehicle document Q&A. Built as part of Datakrew’s technical assessment using LangChain and FastAPI.

View Repository

Key Features

  • FastAPI backend for tool chaining and streaming
  • Agent routing logic integrated with LangChain
  • Embedding retrieval via pgvector and PostgreSQL
  • Minimal React frontend with authentication
  • OpenAI API integration for GPT-4 output generation

Deliverables

  • Backend API with toolchain management endpoints
  • Secure chat and embedding interfaces
  • Frontend interface for document interaction
  • Containerised deployment with environment configuration

Technology Stack

FastAPILangChainpgvectorPostgreSQLOpenAI APIReact
Data Analytics
Completed1 week

Telehealth Data Analytics

Patient Journey Analytics & Targeted Marketing Insights

Completed a comprehensive data analytics assessment for WhiteCoat, applying SQL, Python, and data visualisation to enhance patient engagement, consult flow efficiency, and medication adherence.

View Repository

Key Features

  • Analysed drop-off rates and delays across WhiteCoat’s virtual consult funnel
  • Segmented users by insurance status, delivery preference, and chronic behaviour patterns
  • Identified leading chronic and acute diagnoses, and prescription trends
  • Generated actionable recommendations for product, marketing, and operations

Deliverables

  • SQL queries for diagnosis and medication insights
  • Funnel delay analysis using Python and pandas
  • User segmentation with visualised behavioural insights
  • Executive summary with strategic recommendations
  • Live in-person presentation of findings

Technology Stack

SQLPythonPandasSeabornMatplotlibExcelPowerPoint
Research
Completed1 year

Final Year Project (NTU)

An Empirical Study of Convolution-based vs Transformer-based Diffusion Models

A comparative study exploring the performance and inductive biases of convolutional (U-Net) and transformer (DiT) architectures in diffusion models for image generation.

View Repository

Key Features

  • Implemented DDPM and DiT-S/2 from scratch using PyTorch
  • Introduced Frequency-based Noise Control (FNC) as a novel inductive bias
  • Logged training metrics using TensorBoard and Weights & Biases
  • Evaluated FID, IS, PSNR, and SSIM to assess model performance
  • Analysed scalability across architecture size and training complexity

Deliverables

  • Final research report with experimental results and visual analysis
  • Reproducible PyTorch training and evaluation scripts
  • LaTeX-based thesis document with diagrams and references

Technology Stack

PyTorchTensorBoardLaTeX
Interested in working together? Let's create something amazing!
Professional Credentials

Certifications

Professional certifications and credentials that validate my expertise in modern technologies and development practices.

×
Verified

Google Cybersecurity

Provider:Google
Platform:Coursera
Issued:November 2023

Skills Validated

Cybersecurity Incident ResponseThreats, risks, and vulnerabilitiesSecurity frameworks and controls
×
Verified

Google Business Intelligence

Provider:Google
Platform:Coursera
Issued:September 2023

Skills Validated

Big QueryBusiness AnalysisDashboard ReportingETLTableauData ModellingData Visualisation
×
Verified

Google Advanced Data Analytics

Provider:Google
Platform:Coursera
Issued:March 2022

Skills Validated

Data ScienceKaggleMachine LearningStatistical AnalysisTableauEDAPredictive Models
3 Professional Certifications
Verified by Industry Leaders
Contact

Let's Work Together

Ready to bring your ideas to life? I'm always excited to work on interesting projects and collaborate with amazing people. Let's create something extraordinary together.

Or reach out directly: