Dr Mohammad Heydari

From Academia to Application: Securing the Future with AI Wisdom Practical AI Solutions for Real Impact

Email: info@heydari.info
Phone: +44 78 044 99 044

Introduction

Dr Mohammad Heydari is a cyber security academic and researcher with hands-on experience turning advanced research into practical, deployable security capabilities. He is currently MSc Cyber Course Leader & Lecturer in AI & Cyber Security and has contributed to a number of national and EU funded projects in the field of AI/ML enabled solutions in Cyber Security.

What he helps you to achieve:

  • Reduce cyber risk using AI/ML-driven detection, threat intelligence, and security-by-design
  • Deploy GenAI safely (governance, risk controls, evaluation, red-teaming mindset)
  • Modernise SME services by identifying high-ROI AI use cases and taking them from idea → pilot → rollout
  • Deliver complex projects in the field of AI/ML with clear scope, stakeholders, milestones, and measurable outcomes

Academic

Courses (September 2025)

  • Data Analytics
  • Advanced Topics in Cyber Security
  • Machine Learning in Cyber Security
  • Quantum Machine Learning Lab

Recent Talks

Consulting Services

  • AI Readiness Assessment (data, people, process, cyber risk, compliance)

  • Threat detection & anomaly detection using Discriminating/Predictive AI (design, evaluation, prototype implementation)

  • AI-enabled threat intelligence (CTI pipelines, supply-chain risk insights)

  • Security testing with a domain-tuned “security tester” LMs (Responsible evaluation frameworks; Exploit-generation research insights)

  • IoT & supply-chain security (Graph mining aware solutions for asset modelling, attack-path simulation and resilience planning)

  • AI Readiness Assessment (data, people, process, cyber risk, compliance)
  • Use-Case Discovery (Design Thinking): problem framing → ideation → prioritisation → rapid prototyping (focused on measurable ROI)
  • Knowledge & Workflow Assistants
    • RAG for secure enterprise search over your internal documents and knowledge base
    • MCP-based integrations to connect agents to your tools/data (ticketing, CRM, databases, internal services) in a more standardized way
  • Developing / Customising Language Models for your SME
    • Domain adaptation (instruction tuning / fine-tuning where appropriate), prompt + policy layers, evaluation, and guardrails
    • “Your Business Language”: terminology, products, policies, and workflows baked into the assistant
  • Multi-agent solutions (when one model shouldn’t do everything): e.g., Security Analyst Agent, Compliance Agent, Data Agent, Ops Agent coordinated with clear permissions and auditability (MCP tool access + RAG context)
  • Governance & Assurance: Risk register, Evaluation metrics, Human-in-the-loop, Model monitoring, Secure deployment patterns

  • Model a digital twin of your service, process, or security posture

  • Run “what-if” scenarios before production investment (changes, incidents, demand spikes, control/control failures)

  • Use the twin for risk-aware decision support and prioritization

  • Layer GenAI assistants on top of the twin to:

    • Explain trade-offs and recommended actions

    • Generate executive/technical reports

    • Guide incident response and operational playbooks

  • Leverage simulation/test-bed style evaluation to assess attack paths and detection performance in a practical, business-ready way

Recent Projects

September 2025 – now

Innovate Funded Project: SECURE-AI

  • SECURE-AI empowers SMEs to understand AI-driven cyber threats while also equipping them with practical, AI-enabled defense strategies.
  • The project further fosters a sustainable ecosystem by connecting SMEs with cyber security specialists and developing long-term exploitation and community-building plans.

2020 – 2022

EU Cyber Security Funded Project: CYRENE (Supply Chain Cybersecurity)

  • Contributed to the EU-funded Cyrene project (EU Horizon 2020), focused on developing a conformity assessment method for evaluating supply chain services in the context of cybersecurity.
  • Secured funding contributions and led research on:
    • Designing an ML-based testbed for simulating attack paths.
    • Developing a distributed ML-based method for attack detection.
  • Collaborated with industry partners to translate research findings into practical cybers ecurity solutions.

2019 – 2021

EPSRC Standard Grant: FaCT – Faithful Composition of Trust

  • Led the design of a formally verified remote attestation protocol on top of seL4.
  • Developed lightweight protocols relying on seL4’s guarantees and designed end-to-end security frameworks for communicating seL4-TEE systems.