I'm an AI Engineer at insightfactory.ai, where I build and scale multi-agent systems. I studied Computer Science at the University of Adelaide, and previously worked as a Machine Learning Engineer at the Australian Institute of Machine Learning, where my research focused on adversarial and backdoor attacks against state-of-the-art object detectors, which got published at top-tier security conferences.
I write about scalable ML, AI systems, and things I find interesting at harryng.dev.
- Designing production multi-agent architectures on autoscaling, fault tolerance, observability
- RAG infrastructure with Qdrant, PostgreSQL, and custom ingestion pipelines
- Writing about distributed systems patterns applied to AI agent infrastructure
- Contributor to vercel-labs/just-bash (optimizing grep to be 100x faster)
Always happy to talk about harness engineering, distributed systems, agent infrastructure, ML security and Deep neural nets.