Dialogue
I’m currently pursuing a Master of Information Technology in Artificial Intelligence at UNSW, with experience across production data pipelines, distributed systems, LLM applications, and ML infrastructure.
Character Profile
This section makes the profile clear for recruiters while keeping the game-style personality.
I build software at the intersection of backend engineering and artificial intelligence. My experience includes production data pipeline monitoring, distributed ETL systems, LLM-powered analytics, and scalable model-serving APIs. I care about clean architecture, observability, reliability, and building products that feel simple for users.
Education
Master of Information Technology in Artificial Intelligence at UNSW.
Current Role
Software Engineering Intern building pipeline monitoring and observability systems.
Target
Software engineering internships in backend, infrastructure, AI systems, or data platforms.
Quest Log
Quest 01 · Current
Python · AWS · Data Pipeline Monitoring · Observability
Built monitoring for production data freshness, pipeline health, and ETL failures.
Reduced mean time to diagnose pipeline issues by 60% through dashboards and alerts.
Implemented anomaly detection to catch failures and upstream data issues earlier.
Quest 02 · Research
PyTorch · Diffusion Transformers · Model Optimization
Implemented DiT super-resolution pipelines for high-resolution image reconstruction.
Reduced inference latency using pruning, quantization, and knowledge distillation.
Achieved 4× model compression while maintaining comparable image quality.
Quest 03 · Data Platform
Python Backend · ChatGLM · SQL Analytics · Reliability
Integrated ChatGLM with internal analytics systems for conversational food data analysis.
Converted natural language questions into SQL analytics workflows and visual results.
Improved pipeline reliability by 35% through retry logic and error-handling improvements.
Quest 04 · Scale
Distributed ETL · IoT Telemetry · FastAPI · SQL
Built distributed Python ETL pipelines for large-scale IoT telemetry ingestion.
Optimized SQL queries and batch processing logic for throughput and reliability.
Developed a FastAPI QA service, improving throughput by 24% and recall by 18%.
Project Inventory
Distributed Systems
A C++ distributed key-value store with a multithreaded TCP server, peer replication, and thread-safe request handling.
ML Infrastructure
A containerized LLM inference service using FastAPI, Docker, Kubernetes autoscaling, monitoring, and MLflow versioning.
LLM Applications
A retrieval-augmented generation pipeline for document summarization with semantic search and low-latency FastAPI serving.
Skill Inventory
Python, C++, Java, SQL
FastAPI, API Design, Linux, Concurrency
AWS, Docker, Kubernetes, Redis, PostgreSQL
PyTorch, MLflow, LLM Apps, Kafka, Airflow