Physics PhD Student @ National University of Singapore (NUS)
I build tools to automate the loop of scientific discovery and complex workflows.
- Auto-Research In Few-lines (arif): A lightweight Python micro-framework for LLM-driven research experiments.
- Motivation: To help users (and AI agents) autonomously construct and monitor customized automated research loops with minimal code, striking a balance between creative freedom and experimental control.
- Key Tech: Python, LLM Automation, Experiment Tracking.
- AgentCommander: An advanced, graph-based workflow execution engine for OpenCode/Claude/Gemini/Qwen CLI.
- Motivation: I built this to automate my own research experiments and model optimization loops.
- Key Tech: Python, LLM Agents, Symbolic Regression.
- PARIS Monte Carlo Sampler: An efficient adaptive importance sampler for high-dimensional multi-modal Bayesian inference.
- Motivation: To tackle complex posteriors with far fewer likelihood calls than conventional methods (e.g., Dynesty/PTMCMC) by elegantly combining global exploration with local adaptation.
- Key Tech: Python, Bayesian Inference, Importance Sampling, High-Performance Computing.
- Languages: Python, C++
-
- Research: Astrophysics, Bayesian inference, Machine Learning
- Interests: AI Agents, Quantitative Finance, High-Performance Computing
Let's connect and build something cool!