Principal AI Engineer · Sydney, Australia
I build AI systems for production, bringing more than two decades of software engineering, security, telecommunications and banking experience to the work. My focus is practical: useful AI products, reliable agent platforms and secure systems that hold up beyond a demo.
Personal website · LinkedIn · Email
- Production AI at Quantium: leading engineering for AI products and enterprise agent platforms.
- AI-assisted software delivery: building reusable skills and workflows for discovery, implementation, evaluation, review and operation in Codex and Claude Code.
- Tools for AI engineers: exploring model routing, coding-agent costs, subscription usage monitoring and retrieval evaluation.
- AI and psychology research at Macquarie University: pursuing a Master of Research, combining physiological, vocal, behavioural and audience signals to understand live speech performance and support practical feedback for speakers.
| Project | What it does |
|---|---|
| AI-SDLC | A pack of 20 skills, templates and supporting research for AI-assisted software delivery in Codex and Claude Code. |
| Usage Monitor | A macOS menu bar app for monitoring usage limits across AI subscriptions and coding tools. |
| Model Switcher | Per-prompt model routing and offline cost tracking for Claude Code. |
| Claude Session Analyser | Local session analytics, token-saving insights and cost forecasting for Claude Code. |
| MyAutoRAG | A command-line toolkit for building and evaluating retrieval-augmented generation pipelines from PDFs. |
| MCP Examples | Practical examples of connecting agents to maths and weather tools through the Model Context Protocol. |
Before Quantium, my work at Telstra spanned production AI, cloud platforms and payments:
- AskTelstra: principal engineering and architecture improvements for a production AI application on Azure, built with Microsoft ISE, making AI queries 5× faster.
- API platforms: migrating more than 30 services from Apigee to an internal API gateway without customer downtime.
- Payments: modernising prepaid recharge and payment platforms supporting more than 1.5 million daily transactions.
Across these systems, I care about secure architecture, evaluation, observability and making software easier to operate.
- AI engineering: LLM applications, agentic systems, RAG, MCP, model evaluation and AI-assisted development.
- Languages: Python, Java, Swift and SQL.
- Platforms: Azure, AWS, Docker and Kubernetes.
- Security and delivery: threat modelling, secure APIs, CI/CD and production monitoring.
Alongside my current research at Macquarie, I've completed a Master of Information Technology in Data Science, Engineering and AI and a master's degree in Cyber Security at UNSW, an MBA at RMIT, and a Bachelor of Engineering in Computer Engineering at Saurashtra University.
I also mentor engineers and students, sharing lessons from building systems, navigating technical careers and continuing to learn while working.
I'm a runner with 14 marathons across four continents behind me. Research, mentoring and running all give me reasons to stay curious and keep improving.
I'm always happy to compare notes on AI engineering, secure software, open source and mentorship.
- Website: jiggykakkad.com
- LinkedIn: linkedin.com/in/jiggy
- Email: jig2nesh@gmail.com