Australia
I am an AI software engineer with 8+ years of experience across AI and software engineering, specialising in building and deploying ML-powered systems in production. Experienced across the full ML lifecycle, including data pipelines, model development, evaluation, deployment, and monitoring. Hands-on experience with LLM-based systems (e.g., RAG, AI agents) and a strong focus on improving reliability, safety, and real-world performance of AI applications. Passionate about building scalable platforms that enable teams to safely and efficiently ship AI-driven features.
Built and deployed an LLM-powered support agent using Retrieval-Augmented Generation (RAG) to assist engineers in resolving customer issues. Developed evaluation framework to measure answer accuracy and iteratively improved solution rate from 10% to 32%. Optimised prompt engineering strategies and retrieval logic to reduce hallucinations and improve reasoning performance. Established feedback loop for continuous model improvement based on real usage data. Designed and maintained scalable serverless infrastructure (AWS Lambda, SQS, DynamoDB) to support real-time ML-powered services and notification delivery. Built APIs for model inference and integrated LLM-based services into production workflows. Implemented monitoring and observability pipelines to track latency, failure rates, and system reliability across distributed services.
Penultimate year PhD student in UTS. • Working in Australian Artificial Intelligence Institute (AAII) in UTS. My research areas cover data science, data mining, machine learning, etc. • Computer science and mathematical/statistical background as a Data scientist with 3 years of experience. Skills include Data science, Machine Learning, Statistics, Data warehouse, Cloud, Python, Kubeflow, Java, Tensorflow, PyTorch, AWS, SQL etc. • Preprocessing and cleaning data, performing analysis using machine learning techniques. • Presenting outcomes in a clear manner. • Received ARC Linkage Scholarship and International Research Scholarship. • Selective paper published: Semi-supervised Variational Multi-view Anomaly Detection Accepted by Core Conference • Reviewer of tier-1 machine learning conferences (Neurips, AAAI, etc). • Stakeholder management skills.
• Developed state-of-the-art outlier detection algorithm based on a variant of Variational Autoencoder in python and Tensorflow in collaboration with Prof. Makoto Yamada in Riken-AIP. • Preprocessing and cleaning data, performing analysis using machine learning techniques. • Presenting outcomes in a clear manner. • Relevant research paper had been submitted to top conference in machine learning.
• Tutored Machine Learning & Data Mining in UNSW, communicated and collaborated with lecturer, students and other tutors to address corresponding issues. • Developed assignments and tutorial material in python.
• High Distinction in following courses: Machine Learning and Data Mining Extended design and analysis of algorithm Big Data Management System Capacity Planning • Distinction in following courses: Data warehouse and Data mining Deep Leaning and Neural Network Computer Vision • Participated in research of natural language process in Prof. Wei Wang’s group. • Tackled complex problems of name entity recognizer on twitter by machine learning approach. • Deployed algorithm in Java on AWS (S3, EC2). • Top 10 Performance in VIVA hand detection challenge. Jun 2017 • Won 3rd place in Microsoft College Code Competition. Mar 2017 • 6th place in Jane Street Electronic Trading Competition. Mar 2018