Application Security & Security Research · PhD in Information Security
Secure code review · secure SDLC · web & API security — London, UK

Security researcher and application security practitioner. I focus on finding and fixing web and API vulnerabilities, reviewing source code, and building security into the SDLC — while keeping an offensive edge through OSCP-style lab work.

• Web & API security — OWASP Top 10, authentication / authorization and business-logic flaws; responsible disclosure, including a multi-round technical dialogue with Apple Security.
• End-to-end AppSec case study — secure-llm-sdlc-reference: three security failures — broken object-level authorization, indirect prompt injection / excessive agency and SQL injection — traced from threat model and reproducible vulnerable baseline through hardened fixes, 19 regression tests and CI security gates.
• Secure code review — source-code-security-scanner: my Python triage scanner with severity scoring, entropy-based secret detection, and SARIF / CI build-gating output — for hardcoded credentials, injection sinks, weak crypto and insecure deserialization.
• Secure SDLC / DevSecOps — a GitLab CI/CD pipeline integrating SAST (SonarQube), DAST (OWASP ZAP), container scanning (Trivy) and IaC scanning (Checkov), deployed to AWS (EC2 / VPC / IAM / Security Groups).
• Offensive testing — web exploitation, Linux & Windows privilege escalation, and Active Directory attack paths (OffSec PEN-200 / OSCP preparation).

📌 oscp-methodology — my full-chain penetration testing playbook, written as a field checklist: recon → foothold → privilege escalation → Active Directory.

Currently exploring: securing enterprise AI adoption and AI-assisted security workflows.

PhD in Information Security (Royal Holloway, University of London). My research studied a concrete web privacy threat: the standard JavaScript APIs every browser exposes — keyboard, mouse and mobile motion/orientation events — can be abused by any web page, with no permission prompt, to silently de-anonymise the person behind the browser, defeating private / incognito mode and logged-out browsing.

• Browser attack-surface analysis — mapped which behaviour-leaking APIs are reachable without user consent across the major browsers on Windows, macOS, iOS and Android, in both normal and private modes.
• Threat demonstration — showed how a website (or a third-party script embedded across many sites) can build a persistent behavioural profile of a visitor and re-identify them across later sessions, even when they believe they are anonymous.
• Defensive recommendations — proposed practical browser- and user-side mitigations to put this tracking capability back under the user's control.

📌 browser-behavioural-biometrics — sanitised research code: the data-collection browser extension, the browser-API attack-surface study, and the identification pipeline. Thesis: Browser user privacy — identifying users via browser interactions (RHUL, 2023).

Beyond security, I build race-forensics tooling, real-time telemetry, hardware integration, and a rules engine for a competitive Assetto Corsa league — all league-safe (read-only, no car-file edits, no checksum impact).

Python · JavaScript · PHP · SQL · Bash · Lua · C#  |  Burp Suite · OWASP ZAP · SonarQube · Trivy · Checkov · Nmap · AWS · Git

Certifications: CompTIA Security+ · OSCP (in preparation)

_{Building secure software, breaking insecure software — and making cars go faster.}