Projects

A collection of projects I've worked on, ranging from academic research to personal explorations. If you think it would be useful or interesting to collaborate on a project, please contact me to discuss.

Behavior of Chain-of-thought Monitorability

- Present

Studying how monitor effectiveness changes as the capability gap between monitor and target models widens, with a case study on distinguishing sandbagging from genuine incapability.

Chain of ThoughtSandbaggingInspect AIOpenRouter API

Replicate Internal Coherence Maximization (ICM) from Wen et. al (2025)

Praxis Project Trial. ICM (unsupervised) elicits human concepts from base language models by maximizing mutual predictability and local consistency among concept-related examples.

ReplicationAI SafetyUnsupervised Elicitation

Mini Llama2 (CMU Advanced NLP)

Minimalist Llama2 implementation using PyTorch, exploring LLM components like RoPE, self-attention, and AdamW optimizer.

Large Language Model (LLM)PyTorch

Other Projects

SEACrowd Website

-

Jekyll-based static site from scratch for SEACrowd, grass-root org for Southeast Asian AI research.

JekyllBootstrapJavaScriptSCSS

SportConnect

Event management platform connecting sporty people locally. CodePath Advanced Webdev Full-Stack Project.

TypeScriptReactExpressPostgreSQLTailwindCSSDaisyUIBetterAuth

PWA SeizureSavvy

-

Full-stack Progressive Web App for seizure tracking with predictive warnings using XGBoost and LSTM models, built with React and Flask.

Progressive Web App (PWA)ReactFlaskXGBoostLTSMSQLAlchemy

Ask my Second Brain (Obsidian RAG)

-

AI-powered personal knowledge assistant using LlamaIndex and OpenAI API to query Obsidian knowledge base with RAG architecture.

Retrieval Augmented Generation (RAG)PythonLlamaIndexOpenAI

Synthetic Control Replication and Extension

Replication study of Philadelphia's excise tax impact on beverages using synthetic control methods in R with causal inference.

RrmarkdownCausal InferenceSynthetic ControlReplication

Bayesian Hierarchical Modeling for GP Visit Count Data

Analyze GP visit patterns using Zero-Inflated Poisson models with complete and partial pooling, with Bayesian inference and data imputation.

bayesian modelinghierarchical modelsPyMCmissing data imputationhealthcare analyticsstatistical inference

Text classification using SVM and Naive Bayes

Text classification system to automatically classify notes and assignments using SVM and Naive Bayes with 87% accuracy.

statistical machine learningscikit-learnSupport Vector Machine (SVM)Naive Bayes

GitHub Activity

My GitHub ↗ contributions over the past year. Colored squares represent days with commits.

GitHub contribution chart