Skip to content
View coylemichael's full-sized avatar

Block or report coylemichael

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
coylemichael/README.md

Michael Coyle

SecOps Engineer building security automation, vulnerability management tooling, and applied AI systems for real-world operational workflows.

I work across security engineering, automation, and applied AI. Most of my work is focused on reducing manual security toil: vulnerability management, Defender data, KQL-driven investigation, compliance workflows, and AI-assisted operational tooling.

Recently, I have also been exploring context engineering and LLM workflow design through projects like Drift, a structured handoff system for keeping AI-assisted work coherent across sessions.


🔨 I mostly work with

Python PowerShell KQL TypeScript JavaScript HTML5

Azure Azure OpenAI Microsoft Defender Microsoft Graph FastAPI Pydantic


🧭 Current focus

  • Vulnerability management automation using Defender APIs, Graph API, Python, and Azure OpenAI
  • AI-assisted triage, enrichment, and prioritisation of security findings
  • KQL-based threat hunting and security reporting
  • ISO 27001 compliance automation and evidence workflows
  • Entra ID least-privilege analysis and advisor tooling
  • LLM context continuity, structured handoffs, and agent workflow design

🧪 Applied AI / research interests

I am interested in practical AI systems that improve operational security work.

Areas I am exploring:

  • LLM-assisted security operations
  • Context engineering for AI-assisted development
  • Multi-session AI workflow continuity
  • Structured handoff protocols for agents
  • Vulnerability prioritisation and remediation automation
  • Compliance evidence generation and control mapping
  • Human-in-the-loop security automation

🚧 Projects

A lightweight context-engineering workflow for AI-assisted software work.

Drift uses a structured research → plan → handoff pattern to help AI agents preserve useful context across sessions without carrying forward stale, bloated, or irrelevant chat history.

It is designed for situations where work spans multiple prompts, multiple sessions, or multiple agents.

Focus areas:

  • AI session continuity
  • Context handoff design
  • Agent workflow structure
  • Reducing repeated codebase discovery
  • Improving long-running AI-assisted development tasks

🌱 Currently learning

C

I am also deepening my work around applied AI systems, evaluation methods, and context-engineering patterns for security and software engineering workflows.


📫 How to reach me

LinkedIn


⚡ Fun fact

I used to hold a Scottish powerlifting record. Times change.

Pinned Loading

  1. github-azure-cert-expiry-monitor github-azure-cert-expiry-monitor Public

    Python

  2. audio-reset audio-reset Public

    C

  3. light-weight-screen-recorder light-weight-screen-recorder Public

    Lightweight ShadowPlay-style video recorder for Windows. NVENC hardware encoding, WASAPI audio capture, zero re-encoding on save. Pure C, ~5000 lines.

    HTML 2

  4. drift drift Public

    Drift is a lightweight context engineering system for maintaining continuity across agent sessions.

    3