Ryan Ong Wee Kiat

Software Engineering Intern (AI/CV) • Singapore

• Built reusable ML infrastructure components for experiment tracking, dataset versioning, and model evaluation to support reproducible training and rapid iteration across multiple research pipelines.
• Engineered scalable end-to-end computer vision pipelines for heterogeneous clinical imaging datasets, optimizing data ingestion, preprocessing, model training, and evaluation workflows for production-oriented experimentation.
• Optimized GPU training workflows and data-loading pipelines for large-scale image processing tasks, reducing training bottlenecks and improving computational efficiency for deep learning workloads.

Cloud Dev-Ops Engineer Intern • Singapore

• Automated end-to-end infrastructure health checks and deployment workflows using Ansible Automation Platform and GitLab CI, improving system reliability, operational monitoring, and incident response efficiency.
• Built interactive operational dashboards in Microsoft Power BI integrated with automated data ingestion workflows via Power Automate, reducing manual analysis effort by 10%.

MySQL · Python · Scikit-learn · SQLAlchemy · Plotly · hvPlot · Tableau · Docker

• Database Engineering: Designed a robust MySQL schema and automated database creation/CSV ingestion via SQLAlchemy.
• System Reliability: Deployed the solution using a Dockerized MySQL instance to ensure consistent development and production environments.
• Feature Engineering: Implemented cyclical time encoding and weighted logic to identify high-value anomalies in real-time.

Multimodal ML · Computer Vision · NLP · Audio Processing · ETL · Deep Learning

• Data Collection: Automated large-scale data collection using the YouTube Data API, implementing robust filtering, pagination, and failure recovery.
• Pipeline Architecture: Built an end-to-end multimodal ML pipeline to fuse visual, audio, and textual signals for downstream prediction tasks.

Machine Learning · PyTorch · Data Augmentation

• Algorithm Optimization: Adapted SOTA models like MobileNet V4 through parameter optimization to improve species recognition accuracy.
• System Implementation: Constructed a Deep Learning Neural Network on PyTorch to facilitate high-stakes medical decision-making in rural regions.