I'm a Chief Technology & Security Officer with a background spanning cybersecurity operations and leadership, cloud infrastructure, platform architecture and digital identity.
I enjoy moving between strategy and hands-on engineering: designing secure services, investigating how systems behave, building tools that make complex information useful and experimenting with embedded hardware. This GitHub is where those ideas become working projects.
My approach: Security should be designed in, operational reality matters, and the best way to understand a technology is often to build with it.
| 🔐 Cybersecurity & DFIR | ☁️ Cloud & Architecture | 🛠️ Software & Hardware |
|---|---|---|
| Security operations, telemetry, incident response and digital forensics | Secure-by-design services, identity, AWS, infrastructure and resilient platforms | Linux desktop apps, visualisation, embedded systems, wireless and GNSS projects |
|
A security operations analyst workspace bringing together IOC management, threat enrichment, investigation utilities and incident workflows. |
A receive-only ESP32 wireless intrusion detection prototype, with management-frame telemetry, anomaly rules and a Python collector. |
|
Live network traffic visualised as a 3D city: hosts become buildings, subnets become districts, and conversations become highways. |
A Fedora GNOME desktop tool for live GNSS diagnostics and troubleshooting of a USB-connected u-blox NEO-M9N receiver. |
🎛️ netspectrumA GPU-accelerated, audio-equaliser-inspired visualisation of live network traffic. A playful way to see protocols in motion. |
A Linux CLI for tracking public IP address changes, geolocation data and optional change notifications. |
- Wireless security — passive sensing, authentication telemetry and explainable anomaly detection.
- Digital identity & privacy — trustworthy verification, selective disclosure and secure architecture.
- Cybersecurity tooling — SOC workflows, DFIR and richer ways to interpret telemetry.
- Embedded engineering — GNSS, ESP32, STM32 and Raspberry Pi systems.
- AI-assisted engineering — using AI to speed up delivery without compromising review, security or quality.
For collaboration, technical discussion or to follow what I'm building, explore my public repositories or start a conversation in a relevant project's GitHub Discussions or Issues section (where enabled).



