Enterprise agent platforms
Multi-agent systems for code analysis, incident investigation and developer productivity, with MCP tool access across Grafana, Jira and internal services.
Ownership from architecture to observability.
I'm an AI engineer with production experience building agentic AI infrastructure at enterprise scale — currently deploying multi-agent systems, MCP-based tool integrations and LLM applications on AWS at one of Australia's largest financial services institutions.
I like the messy middle of AI engineering: the part where a demo has to become a service with owners, guardrails, cost control and dashboards. That has meant building reusable CI/CD workflows adopted across 20+ repositories, and wiring MCP servers into enterprise infrastructure so tools can be discovered and invoked dynamically.
A Python-first background with hands-on frontend experience (Next.js) means I can take a workflow from a model call through to the interface someone actually uses — and a research habit that started with Chinese text-similarity models and natural-language-to-SQL keeps me honest about what these systems can really do.
AI observability
Grafana dashboards and telemetry pipelines for evaluating and monitoring production AI applications — so quality is measured, not assumed.
Developer platforms
Reusable CI/CD workflows, repository migration and vulnerability-scanning automation that teams adopt by default.