Researcher, builder, and 12th-grade student based in Singapore. I investigate the limitations of artificial intelligence, develop transparent educational technology, and compete in international informatics olympiads.
This site documents my research, technical projects, and the ideas that shape my thinking.
Subscribe to this blog via RSS feed
Explore by Topic
- Artificial Intelligence in Law – Research on LLM failure modes, "Statutory Myopia," and multi-agent legal reasoning systems.
- EdTech & Democratized Learning – Building transparent, governance-compliant AI tutoring architectures.
- AI Systems Research – Research on LLM Cache Networks, AI Systems.
- Cybersecurity & Informatics – Cryptography, digital forensics, and competitive algorithmic problem-solving.
- Climate & Sustainability – Data-driven modeling for hybrid renewable energy solutions.
- The Library – Reading notes, book reviews, and essays on productivity, management, and literature.
Featured Writings
- Reasoning Over Recall: Why Statute-Only RAG Degrades Performance (12 Apr, 2026)
A plain-English breakdown of why providing statutory text without accompanying case law degrades LLM ensemble performance in legal tasks. - From Semantic Reuse to Auditable Tutoring (07 Jul, 2026)
Proposing a "glass-box" extension of semantic caching to make AI tutoring transparent, Socratic, and fully auditable. - Deep Work — Cal Newport (26 Nov, 2024)
Reflections on cultivating intense focus and eliminating distraction in an increasingly noisy world.