Shihab Sahoriar
Open to opportunities

// curriculum vitae

Shihab Sahoriar

GTFounder of GTRepo

MSc Cybersecurity & Digital Forensics Student

London, United Kingdom
Cybersecurity graduate roles

MSc Cybersecurity & Digital Forensics student at the University of Wales Trinity Saint David, London. Building Zero Trust AI systems, forensic investigation tools, and SOC dashboards.

View ProjectsPDF CV coming laterContactGitHub
Zero Trust ArchitectureAI SecurityDigital ForensicsSOC SystemsMemory Forensics

// about me

Who I Am

MSc Cybersecurity & Digital Forensics student at the University of Wales Trinity Saint David, London Campus, with a background in Computer Science and a strong interest in AI security, Zero Trust Architecture, digital forensics, SOC systems, and secure AI infrastructure. I am currently building practical cybersecurity systems including a Zero Trust AI Gateway, AI Firewall concepts, SOC monitoring dashboards, and memory forensics platforms.

Originally from Gazipur, Dhaka, Bangladesh, I am now based in London where I study and build. GT is my personal technical identity, and GTRepo is the platform where I document systems, research notes, and experiments as my understanding of modern cybersecurity evolves.

My professional goal is to develop into a cybersecurity professional specializing in AI security, Zero Trust systems, SOC monitoring, and secure AI infrastructure while contributing to practical and research-oriented cybersecurity solutions.

Born

18 September 2002, Gazipur, Dhaka, Bangladesh

Based in

London, United Kingdom

Studying

MSc Cybersecurity & Digital Forensics, expected 2026

University

University of Wales Trinity Saint David, London Campus

Email

sihab.vip@gmail.com

GitHub

https://github.com/AIWshihab

Brand

GT (personal) · GTRepo (platform)

// journey

Timeline

  • Postgraduate study at UWTSD London Campus — Current
  • Expected completion: 2026
  • Specialising in cybersecurity, digital forensics, and AI security systems
  • MSc dissertation: Zero Trust AI Gateway for secure AI model serving
  • Research in memory forensics, SOC systems, and explainable security
  • Combining academic rigour with practical system building

// skills

Technical Skills

Learning
Practical Experience
Research Project Experience
Zero Trust ArchitectureResearch Project Experience
AI Security ConceptsResearch Project Experience
Prompt & Output SecurityResearch Project Experience
SOC Monitoring ConceptsLearning
Threat monitoringPractical Experience
Security loggingPractical Experience
Risk AnalysisLearning

Current Learning Focus

SOC Analyst SkillsLinux AdministrationDocker & DeploymentThreat DetectionAI Security SystemsCloud Security Fundamentals

// projects

Featured Projects

Zero Trust AI Gateway

Active — MSc Dissertation

AI Security / Zero Trust

A security gateway for AI model serving built on Zero Trust principles. Inspects prompts, evaluates risk with a policy engine, makes ALLOW / CHALLENGE / BLOCK decisions, forwards safe requests to providers, audits outputs, and produces explainable SOC-style decision traces.

FastAPIReactPostgreSQLDockerZero TrustAI SecurityPolicy Engine

MemScope

Prototype

Digital Forensics

Memory forensics investigation system that ingests forensic datasets and artefacts, normalises process / network / DLL / command evidence, runs correlation rules, and surfaces suspicious process lineages, threat indicators, and attack-chain views.

FastAPIReactVolatility3PostgreSQLDigital ForensicsThreat Correlation
Repository coming soon

AI Firewall / API Interceptor

Experimental

AI Gateway Security

Extension of the Zero Trust AI Gateway for external applications. Supports client API key authentication, prompt inspection, policy enforcement, output inspection, and structured logging with drop-in API integration.

FastAPIAPI GatewayClient API KeysAI FirewallDockerSecurity Middleware

SOC Monitoring Dashboard

Active UI

Security Operations

SOC-style dashboard visualising alerts, user anomalies, attack timelines, threat heatmaps, model posture events, and Zero Trust decision logs. Integrates as a frontend module of the AI Gateway backend.

ReactTypeScriptMonitoringDashboardSecurity AnalyticsThreat Intelligence

Aegis — Personal AI Assistant

Concept

AI Assistant / Automation

Local-first personal assistant concept with voice control, secure tool access, memory persistence, and privacy-aware AI orchestration — designed to run without third-party data sharing.

FastAPIReactVoice AssistantLocal AIAutomationSecurity
Repository coming soon

// experience

Professional Experience

Chef / Kitchen Team Member

London Hospitality Sector

Part-time / London

Worked as part of a professional kitchen team in London's hospitality sector while studying. This role developed strong operational discipline, real-world team experience, and pressure management — all of which transfer directly to structured, high-stakes technical environments like cybersecurity operations.

Transferable Strengths

Working under pressure

High-volume, fast-paced kitchen environments requiring consistent output.

Fast decision-making

Adapting to changing demands and priorities in real time.

Team coordination

Coordinating roles and responsibilities across a kitchen team.

Time management

Managing multiple simultaneous tasks to precise timelines.

Operational consistency

Maintaining quality standards under high load and stress.

Operational discipline

Following processes precisely — a discipline directly applicable to security work.

// research

Research Interests

Zero Trust Architecture

Applying never-trust-always-verify to AI model serving and API access control.

AI Security

Securing LLM systems, AI gateways, and model inference endpoints against misuse.

Secure AI Gateways

Policy-enforced middleware for AI provider routing, prompt inspection, and output filtering.

SOC Monitoring

Security operations systems for real-time event correlation and threat visualisation.

Threat Intelligence

Identifying and correlating threat indicators across forensic evidence and logs.

Explainable Security Systems

Making security decisions transparent, logged, auditable, and human-readable.

Digital Forensics

Memory forensics, artefact processing, and evidence-based investigation pipelines.

Secure Model Infrastructure

End-to-end security architecture for AI model serving environments.

Human-readable Security Dashboards

Presenting complex threat data clearly to SOC analysts and researchers.

// beyond the terminal

Personal Interests

Travelling

Exploring cities, cultures, and new visual perspectives

Movies

Cinematic storytelling, visual craft, and narrative structure

Anime

Character-driven narratives, visual storytelling, and world-building

Series

Long-form narrative across drama, thriller, and science fiction

Technology

Experimenting with new systems, tools, and ideas

Cybersecurity Research

Reading, following the field, and building in the space

Visual Storytelling

Combining technical craft with creative and aesthetic expression