Syed Ali Turab, CISSP | AI Security Engineer & Vulnerability Management Architect

01 / About

I have 7+ years of experience in cybersecurity and AI, spanning hands-on security engineering, detection and response, vulnerability management at scale, and applied machine learning for threat detection and risk prioritization.

My education includes a Bachelor of Applied Science (Honours) in Cyber Security from Sheridan College. I hold CISSP. I'm completing a Master of Management in Artificial Intelligence (MMAI) at Queen's University Smith School of Business, where I apply AI directly to the security problems I work on every day.

My work has covered Data Loss Prevention, Digital Forensics, Cloud Security, SIEM & Detection Engineering, Firewall & Network Architecture, Security Consulting, and Vulnerability Management — from detection engineering and DFIR to enterprise consulting and leading L3 vulnerability management programs. On the AI side, I focus on LLM security, prompt injection detection, anomaly detection pipelines, NLP, and multi-modal AI systems, building tools and research at the intersection of AI and cybersecurity.

I'm also a leader in the community: I serve as Board Director at ISC2 Toronto Chapter (Toronto, ON), leading professional development for one of Canada's largest cybersecurity communities, and I teach Digital Forensics and Cloud & IoT Security at Seneca Polytechnic (Toronto, ON) — my students have rated me on Rate My Professors. I'm a blockchain enthusiast and stay active on-chain — find me at turab.eth and turab.sol.

I'm currently pursuing the OSCP — because the best defensive engineers understand how attackers think. My work combines deep enterprise security engineering (vulnerability management, container security, DevSecOps automation) with applied AI research: LLM prompt injection tooling, anomaly detection pipelines, and reinforcement learning for adversarial attack simulation. I build things, teach what I know, and stay close to where the threats are moving.

03 / Experience

Royal Bank of Canada 2025 – Present

Lead Cyber Security Engineer · Toronto, ON

L3 technical lead for enterprise vulnerability management across cloud, container, and on-prem environments.
Own and govern enterprise scanning platforms, improving scan success rates and asset discovery accuracy by 30–40%.
Designed and automated large-scale scanning and remediation workflows, reducing manual effort by 50%+.
Integrated AI/ML techniques into security workflows for anomaly detection and predictive risk prioritization.
Deployed MCP servers via Docker for the team and actively drive AI adoption to enhance security workflows across endpoint, cloud, and container environments.
Embedded security scanning into CI/CD pipelines to enforce shift-left security at scale.
Oversee vulnerability governance across multiple global subsidiaries.

Teaming Labs 2025 – Present

Lead AI Engineer · Amsterdam, North Holland, Netherlands

Lead architect for a multi-modal AI system designed for collaborative simulation environments.
Implemented dynamic voice and text routing with context-aware tone and emotional adaptation.
Advanced prompt engineering and multi-modal orchestration for high-realism AI interactions.

ISC2 Toronto Chapter 2025 – Present

Professional Development Director · Board of Directors · Toronto, ON Volunteer

Board Director leading professional development strategy for one of Canada's largest ISC2 chapters.
Designing and delivering mentorship, certification, and career programs impacting 500+ cybersecurity professionals.

Seneca Polytechnic 2024 – 2025

Professor · Digital Forensics & Cloud Security · Toronto, ON

Delivered Digital Forensics and Cloud & IoT Security courses to cohorts of 40+ students per term.
Mentored students into cybersecurity internships, co-ops, and entry-level roles through applied instruction.

IBM 2022 – 2025

Security Specialist · Markham, ON

Delivered client-facing enterprise cybersecurity across government, healthcare, and financial services sectors.
Collaborated on AI-based anomaly detection prototypes for security monitoring.
Automated patch-management and coordination workflows, reducing manual effort by 30%+.

Canadian Imperial Bank of Commerce 2019 – 2022

Detection Engineer · Senior Information Security Analyst · Toronto, ON

Started as a co-op student, was brought back part-time in a lead capacity, and upon graduating in November 2020 was promoted full-time as a Detection Engineer.
Built and tuned SIEM/SOAR detections reducing SOC false positives by ~35%.
Managed 200+ EDR detection rules; automated IOC ingestion and enrichment pipelines.
Led DLP and insider-threat detection initiatives; contributed to red-team exercises.

04 / Projects

Multi-Modal AI Role Player for Collaborative Decision-Making

Lead AI Engineer · MMAI Capstone · Teaming Labs, Amsterdam

Designed and engineered an NLP-driven, context-aware AI role player for team-based decision-making simulations — built in partnership with Teaming Labs. The system uses LLMs with advanced prompt engineering to dynamically adapt tone, pacing, and response style in real time. Agentic workflow logic enables the system to reason over conversational state, generate structured outputs aligned to scenario objectives, and route between voice (TTS) and text modalities contextually. The core challenge was translating ambiguous team dynamics and real-world business problems into a reliable AI interaction — requiring tight feedback loops between system design and live simulation testing. Improved participant engagement and decision quality across scenario exercises.

NLP Agentic Workflows Multi-Modal AI TTS Prompt Engineering Structured Outputs

AI Security Research Platform

Security Tooling & Research · Personal

AI-enabled security research and tooling platform focused on prompt injection detection, OWASP LLM Top 10 automation, DeFi exploit detection, anomaly detection pipelines, and AI-driven risk scoring systems.

LLM Security Offensive AI AI Risk Engineering OWASP

PetCare Agentic System

AI Developer & Security Assessor · NLP · Agentic Systems

End-to-end AI agent for veterinary triage: symptom intake via free-text input → urgency classification → care pathway routing → structured clinical summarization for practitioners. Built and then stress-tested from an attacker mindset — conducting a targeted security assessment covering: prompt injection (adversarial inputs designed to override triage logic), input manipulation (malformed/ambiguous symptoms causing classification drift), hallucination risk (edge cases producing clinically unsafe outputs without grounding), and validation path gaps (weak boundaries allowing unsafe content to propagate). Built and assessed from both a developer and attacker mindset — AI in high-stakes domains needs security thinking from day one.

NLP Agentic Systems AI Triage Design Prompt Injection AI Security Assessment

github.com/FergieFeng/petcare-agentic-system

LLM Prompt Risk Scanner

AI Security Research · Personal

Automated scanner for detecting prompt injection vulnerabilities in LLM-powered applications. Maps findings to OWASP LLM Top 10, providing structured risk scoring and mitigation recommendations for enterprise AI deployments.

LLM Security Prompt Injection Risk Scoring

github.com/turaab97/LLM_OWASP_Scanner

DeFi Exploit Detection

Blockchain Security · AI / Anomaly Detection · Personal

Built ML-powered tooling to detect exploit patterns and anomalous transactions across DeFi protocols. Applied anomaly detection and pattern recognition to on-chain data to surface potential attack vectors — including flash loan exploits, reentrancy patterns, and unusual liquidity movements. Bridges blockchain security with applied AI, aligned with CBSP certification and active on-chain presence (turab.eth / turab.sol).

Web3 / DeFi Blockchain Security Anomaly Detection Python

github.com/turaab97

IoT Forensics Traffic Simulation

Python · Scapy · Academic

Used Scapy to simulate MQTT traffic from virtual smart home devices for forensic analysis. Exported traffic to .pcap for Wireshark — no physical IoT devices needed.

Python IoT Wireshark

github.com/turaab97/MQTT_IoT_Scapy

IoT Forensics Investigation Framework

Digital Forensics · Research

Forensic investigation framework for IoT and container environments. Covers evidence collection, chain-of-custody workflows, and artifact analysis across embedded Linux, mobile, and cloud-hosted IoT deployments.

DFIR IoT Security Container Forensics

github.com/turaab97

RL for Offensive Security Simulation

In Progress

RL Research · MMAI 845 · Queen's University

A reinforcement learning agent that simulates an attacker navigating a corporate network to reach high-value AI infrastructure — LLM servers, vector databases, model repositories. Built on NASim (Network Attack Simulator, Gymnasium-compatible), the agent learns optimal attack paths across thousands of simulated episodes using a discrete action space: scan → exploit → privilege escalation. Observation space encodes per-host attributes (discovered status, service info, access level, compromise status) plus a global detection score. A custom stealth-aware reward wrapper models a sophisticated adversary that trades impact for evasion — the episode terminates when cumulative detection risk exceeds a threshold, forcing the agent to balance stealth and speed. Comparing PPO and DQN via Stable Baselines 3 across baseline and stealth-aware reward conditions. Output: which hosts serve as critical lateral movement stepping stones, which network changes reduce agent success, and how attacker behavior shifts under different detection regimes — directly supporting AI infrastructure segmentation and monitoring decisions.

Reinforcement Learning NASim PPO / DQN Stable Baselines 3 Adversarial AI AI Infrastructure Security

github.com/turaab97

06 / Skills

Enterprise Security & Platforms

Tenable (SC, IO, Cloud)
Aqua Security (OCP, Azure, AWS, GCP)
Palo Alto Xpanse & NGFW
Fortinet Firewall / WAF
Forescout NAC
FireEye NX / Trend Micro
SIEM/SOAR (ArcSight, Sentinel, Devo, FortiSOAR)
MITRE ATT&CK / Cyber Kill Chain
Proofpoint / Burp Suite
OpenShift & Kubernetes Security
DFIR · IoT & Container Forensics
Cloud Security (AWS, Azure, GCP)
Blockchain Security · DeFi Exploit Analysis
On-Chain Forensics · Smart Contract Auditing

AI & Applied Machine Learning

LLM Security & Prompt Injection (OWASP LLM Top 10)
Reinforcement Learning (PPO, DQN, NASim)
Stable Baselines 3 · Gymnasium
Adversarial AI & Attacker Simulation
Agentic Systems & Workflow Orchestration
NLP · Intent Classification · Structured Outputs
Anomaly Detection Pipelines
Transformer Fine-Tuning (DistilBERT, LoRA)
Embeddings & Vector Search
Multi-Modal AI · TTS / Voice Models
AI Security Assessment & Risk Modeling
IBM Watsonx AI / BigQuery ML

Automation & DevSecOps

Python (APIs, Multithreading, ML pipelines)
PowerShell & Bash
Jenkins & Ansible
MCP Server Deployment (Docker)
CI/CD Integration · Shift-Left Security
DevSecOps Pipelines (container + image scanning)
Prompt Injection Testing & AI Input Validation
Hallucination Risk Analysis (LLM outputs)
KQL / LINQ / Regex / SIGMA Rules
Elasticsearch · ServiceNow API · JIRA API
Web3 & Solidity · CBSP Certified
Git & Version Control

Leadership & Strategy

L3 Technical Leadership
Mentorship & Onboarding
Academic Instruction
ISC2 Board Leadership
Cross-Functional Communication
Enterprise Risk Reporting
Subsidiary Governance
Program Design at Scale
Agile Project Management
AI Innovation Strategy
Stakeholder Alignment
Regulatory Compliance

Syed Ali Turab CISSP

Measurable outcomes at enterprise scale

Educational content

CONINT 2020: OSINT Search Party CTF — My Approach

MEDIUM POSTS

Active research areas at the AI–security boundary

Enterprise Security & Platforms

AI & Applied Machine Learning

Automation & DevSecOps

Leadership & Strategy

Volunteer & professional community

Let's connect