About
I'm Quy-Anh Dang, a researcher based in Hanoi, Vietnam, currently working full-time with Knovel Engineering in Singapore. My background is in applied mathematics and data science, and I'm particularly interested in how AI systems can be made safe, efficient, and practical enough to deploy responsibly in the real world — a thread that runs through the research and engineering work below.
1 Research Focus
My research centers on three core areas: (1) AI Safety and Alignment – developing trustworthy AI systems through red teaming, adversarial robustness, and model alignment techniques; (2) Data-Efficient Learning – designing algorithms and training strategies that achieve strong performance on small, high-quality datasets under low-resource constraints; and (3) Speech Processing – exploring text-to-speech synthesis and speech recognition systems. I am particularly interested in the intersection of these areas, investigating how to build safe, efficient, and practical AI systems that can be deployed responsibly in real-world applications.
2 Education
Master of Science in Data Science, VNU University of Science — Hanoi, Vietnam
2023 – 2025 · GPA 3.95/4.0 · Class Rank 1st · Department Rank 1st
Bachelor of Science in Applied Mathematics and Computer Science, VNU University of Science — Hanoi, Vietnam
2019 – 2023 · GPA 3.71/4.0 · Class Rank 1st · Department Rank 2nd of 350+ students
3 Work Experience
Researcher, Knovel Engineering
Jan 2024 – Present · Full-time · Singapore
- DeepAssure Platform: Serve as core research team member developing DeepAssure, an AI governance and trustworthy AI platform, from proof-of-concept through initial deployment — implementing core algorithms and optimizing platform infrastructure. In subsequent stages, act as technical expert providing consultation, support, and independent investigation for client AI assurance assessments, with success cases including HTX Singapore Police AI Chatbot, CDL, EarlyBird, and NTU.
- RySense AI Infrastructure: Design and implement production-ready RAG-based AI infrastructure for RySense, Singapore's leading social research organization. Enhance their intelligent survey platform with advanced natural language understanding and automated insight generation capabilities.
- Research: Conduct research spanning AI safety and alignment (red teaming, adversarial robustness, model alignment), data-efficient learning (training strategies for small, high-quality datasets under low-resource constraints), and speech processing (text-to-speech synthesis and speech recognition).
- Engineering: Build data collection and annotation pipelines to support model training and evaluation. Develop PoC applications integrating frontier models with LlamaIndex/LangChain for document processing, and train/deploy optimized open-source LLMs for structured data querying (CSV/Excel). Maintain MLOps evaluation pipelines using Weights & Biases, MLflow, and TrueLens.
Data Scientist, Wingsmob Studio
Jun 2023 – Sep 2024 · Hanoi, Vietnam
- Design and implement automated ETL pipelines to extract data from Adjust API, store in Amazon S3, and load into BigQuery via Google Cloud Functions, processing 1M+ daily events.
- Develop dynamic, real-time dashboards using Power BI to monitor user engagement metrics, resulting in 50% reduction in user dropout rates through data-driven UX optimizations.
- Build and deploy predictive ROAS (Return on Ad Spend) estimation model achieving <5% error rate, enabling optimized marketing budget allocation and improving campaign ROI by 30%.
- Collaborate with product and marketing teams to translate business requirements into analytical solutions and actionable insights.