Cloud & Platform Engineering Lead at hivo.co in Perth, WA. I build infrastructure that doesn't fall over at 3am.
Right now I'm focused on AWS infrastructure, Kubernetes orchestration, and figuring out how to make LLMs useful in production (spoiler: it's harder than the demos make it look).
Day-to-day tech:
- Cloud: AWS (certified 4x), Azure, Terraform, CDK
- Containers: Kubernetes, Docker, ECS
- Code: Python, TypeScript, Go, Shell
- Data: PostgreSQL, Redis, data pipelines, analytics platforms
- AI/ML: PyTorch, LangChain, working with various LLMs
Previously at Insurance Commission of WA and other places in Perth and Singapore. B.Sc Computer Science from NUS.
I curate resources for stuff I actually use:
-
awesome-biostatistics - Clinical trials & medical research tools (300+ resources)
- Started this because I needed to learn the space and couldn't find a good list
- Now includes regulatory compliance, statistical tools, and clinical data standards
-
awesome-n8n - Workflow automation ecosystem (380+ resources)
- Community nodes, templates, enterprise integrations
- Security best practices and hosting guides
-
awesome-vibe-coding - AI coding assistants
- Cursor, GitHub Copilot, Windsurf, and the rest
- Submitted to sindresorhus/awesome
- n8n-mcp-demo - Cross-platform workflow automation with Docker
- daily-learning - Tracking my 30-day contribution streak (currently 23/30)
- AWS data platforms - Production infrastructure for analytics and ML
π₯ 23 days of daily contributions (goal: 30) π 28 PRs open across different projects β 670+ repos starred (probably too many) π¬ 25 issue comments - trying to be helpful, not spammy
Mostly documentation, infrastructure code, and tooling improvements. Recent stuff:
- Fixed awesome-list compliance for DynamoDB resources (PR to Alex DeBrie's repo)
- Apache Arrow doctest bug fixes
- CI/CD workflows for Rust projects
- Kubernetes documentation improvements
- Database backup automation
- How to make Kubernetes not feel like fighting a hydra
- Clinical data standards (CDISC, HL7 FHIR, OMOP)
- Making LLMs work reliably in production
- Rust (slowly but surely)



