Senior Full-Stack Engineer · AI/LLM Systems
I build high-concurrency backends in Java and Go — and, for the past two years, production AI systems on top of them.
Scale, shipped
Full-stack engineer with 10 years architecting high-performance distributed systems — Java and Go services, microservices, data platforms and performance-critical infrastructure.
For the past two years I have focused on production-grade AI: agent architectures, LLM orchestration, RAG, tool calling, monitoring and scalable deployment — paired with modern frontends in React, Next.js and TypeScript, delivered end to end.
In parallel, on Monash’s research pathway, I study blockchain transaction analysis and crypto anti-money-laundering — with two papers accepted in 2026.
Blockchain transaction analysis for anti-money-laundering: finding the right transactions, linking both ends of a cross-chain transfer, and tracing funds through services built to break the trail.
IEEE International Conference on Blockchain (Blockchain 2026)
Systematizes how prior work identifies cross-chain deposits and withdrawals and matches them across bridges, instant exchanges and cross-ledger platforms (2019–2026), compares the evidence each method depends on, and audits whether published datasets and artifacts can still be obtained.
Proceedings of the ACM on Measurement and Analysis of Computing Systems (SIGMETRICS 2027)
Builds MixLaunder, the first public case-level dataset of mixer laundering on Ethereum, and MixGuard, a detection and case-grouping framework evaluated under strict case-level holdout. My part: tracing downstream fund flows after mixer withdrawals, and rebuilding the data-ingestion pipeline.
A modular, LangChain-based framework for enterprise automation.
Production-ready retrieval-augmented generation with hybrid search.
One API in front of multiple LLM providers, with intelligent routing.
Multi-agent systems for automated trading-platform operations.