CS & DS @ Unimelb | Working on ML systems, LLMs, and distributed data processing for AI/ML workloads.
🏆 GSoC 2026 Contributor @ Apache Beam
🏅 ByteDance AI Challenge · Outstanding Project · Top 5/100 · Team Leader
- 🔬 Research Focus: ML systems, LLM infrastructure, and distributed data processing for large-scale AI/ML workloads.
- 🏆 GSoC 2026 @ Apache Beam: Selected for Google Summer of Code 2026 to advance distributed data processing for AI/ML, including streaming ML inference, unbounded sources, and runner internals.
- 🏅 ByteDance AI Challenge: Led a team to the Outstanding Project award, placing Top 5 out of 100 teams.
- 🤖 LLMs & AI Infrastructure: Building and optimizing LLM-powered systems, and contributing to LLM frameworks and AI-native databases.
- 🌱 Open Source: Contributing to Apache Beam, Apache IoTDB (AINode), and LlamaIndex, connecting distributed systems with modern AI/ML.
- 🎓 Unimelb Scholar: Final-year Computer Science & Data Science student.
- ⚡ Fun fact: "It is the time you have wasted for your rose that makes your rose so important."
- 🛡️ CertiScallop: Certified robustness of probabilistic neurosymbolic systems. A logit-margin interface lets existing neural network verifiers (CROWN, global Lipschitz bounds) certify the whole system, evaluated on DeepProbLog, Scallop and ROAD-R benchmarks.
- 🧩 CompoMoLE: Do mixtures of LoRA experts compose skills? A controlled held-out composition diagnostic on HotMoE (AAAI-26) with causal routing analysis.
Apache Beam 
Python & Java SDK · ML Inference · Streaming IO · Runners 
GSoC 2026 Project
|
Apache IoTDB 
AINode · Relational Queries |
LlamaIndex
LLM Framework |
Languages
AI / ML & Deep Learning
Data & Distributed Infrastructure
Web
