Cole Learning Labs · AI Guardian Web Shield · LinkedIn

K–12 technology support by day · cybersecurity senior by night · builder of Guardians in between

I'm a Computer Technical Assistant at Fonda-Fultonville Central School District in New York's Capital Region, finishing a B.S. in Cybersecurity at the University at Albany (SUNY), with a self-designed concentration in AI-Driven Cybersecurity & Ethical Hacking and a minor in Psychology. Students know me as Mx. Cole.

Most of what I build sits where three things meet: defensive security, accountable AI, and people who were never handed a manual. I like systems that explain themselves, interfaces anyone can use, and AI that protects without pretending to be something it isn't.

I build security systems that think before they trust, and explain before they alarm.

• Taking computers apart and putting them back together. Enough that I built a guarded virtual PC workbench (18 screws, 8 removable parts, thermal paste included) so students can do it too without a screwdriver.
• Teaching the tech team mindset. "Follow the clues, try a fix, show how you know it worked." Every lab I write starts with a help-desk ticket, not a lecture.
• Defensive security puzzles. TryHackMe rooms, log review, phishing dissection, and designing honeypots that contain rather than retaliate.
• Philosophy that actually ships. I read moral philosophy and then try to turn it into access policies, audit logs, and UI copy.
• Accessibility as craft. Keyboard paths, reduced motion, plain language, and WCAG 2.2 AA are features, not afterthoughts.
• Community spaces. Gaming stations, a juice bar, and a help desk under one roof is still the dream (see CTRL+J below).
• Building with AI, accountably. I use AI coding agents regularly, review everything they write, and keep a human name on every decision.

Everything I ship is part of one idea: a Guardian protects, explains, and leaves the final call with a human.

flowchart TB
    ethics["AI Ethics & Moral Philosophy<br/>bounded moral learning · accountability · dignity"]
    shield["AI Guardian Web Shield<br/>local-first browser risk guidance"]
    lab["Project Labyrinth<br/>zero-trust access simulator with deception routing"]
    confidant["The Guardian Confidant<br/>privacy-first peer-support companion (design stage)"]
    labs["Cole Learning Labs<br/>100 free K–12 technology activities"]
    ctrlj["CTRL+J Juice Lounge & IT Hub<br/>community technology space (future)"]

    ethics -->|principles| shield
    ethics -->|principles| lab
    ethics -->|principles| confidant
    shield -->|real defensive tooling| labs
    lab -->|zero-trust lessons| labs
    labs -->|free learning + digital literacy| ctrlj
    ethics -->|inclusive by design| ctrlj

A free, browser-based technology learning platform I designed and built for K–12 students, educators, families, and adult beginners. Learners play the school tech team: read a ticket, follow the clues, repair simulated equipment or settings, test the result, and explain whether the original task works.

• 100 activities across hardware and service desk, networking and operating systems, AI literacy and audit, ethical hacking and penetration testing, and a bounded JavaScript/Python coding sandbox.
• Privacy-first by default. Learner work stays on the device. AI literacy labs use authored examples and local rule checks, with no live AI API. Optional cloud replies are off until a learner explicitly turns them on, and even then only an authored scenario ID leaves the browser.
• Accessible on purpose. WCAG 2.2 AA targets, a Display menu for text size, spacing, contrast, and reduced motion, a Listen control, keyboard and touch alternatives for every physical interaction, and structured HTML alternatives for every PDF.
• Built for real classrooms. Grade-band starting points, progressive coaching clues that never do the work for the student, hypothesis-test reports, and a published AI-use notice for district review.

Explore: colelearninglabs.com · View the repository

A published, privacy-first Chrome extension that analyzes webpage, link, and form signals locally and turns them into calm, plain-language risk guidance.

• Recognizes phishing indicators, suspicious links, lookalike domains, risky sign-in forms, and manipulative urgency.
• Connected form-action analysis warns when credentials could leak through a GET submission, an unencrypted destination, or a changed origin, without ever reading field values.
• Zero runtime dependencies, a no-network rule enforced in CI, and honest documentation of what the extension cannot see.

Explore: Install from the Chrome Web Store · View the repository

A defensive Zero Trust simulation that evaluates every access request using identity, role, device trust, MFA, behavior, and anomaly signals.

• Produces allow, deny, or deception-routing decisions and models least privilege, RBAC, and owner-only vault boundaries.
• Routes suspicious sessions into decoy containment with structured audit logging and SOC-style visibility. No hack-back, no retaliation.
• Built with Next.js, React, Tailwind CSS, and SQLite.

Explore: View Project Labyrinth

A living research repository connecting moral philosophy with practical AI governance and security design.

• Develops my proposed framework of bounded moral learning: AI systems that can learn within explicit, human-set moral limits and must escalate rather than improvise beyond them.
• Examines human responsibility, concentrated power, dignity, consent, privacy, accessibility, accountability, appeal, and repair.
• Clearly separates my own philosophical proposals from established standards, and maps each principle onto Guardian systems, Zero Trust, and defensive deception.

Explore: View AI Ethics and Moral Philosophy

A proposed trauma-informed, privacy-first AI peer-support companion. The repository is intentionally design documentation only: a mission, a set of privacy boundaries, and a release gate that must be satisfied with evidence before any build is called safe for sensitive use.

Explore: View The Guardian Confidant

I'm optimistic about AI and skeptical of AI hype, which is why I keep building things that test the difference.

• Local-first, explain-first. Analysis should happen on the user's device whenever possible, and a warning should say why in words a non-expert understands.
• AI must not become a place where humans hide responsibility. Every automated decision I design has a named human owner, an audit trail, and a path to appeal.
• Off by default. Cloud inference, memory, analytics, cameras, and microphones all start off and require an informed, visible choice to turn on.
• Teach the audit, not just the tool. My AI literacy labs ask students to review prompts, claims, bias, and generated code, minimize data, and reject source injection, because the next generation needs to question AI, not just use it.
• Protect and contain, never retaliate. Defensive deception and Zero Trust are about safely saying no, not about striking back.
• Use AI to build, with a human signature. AI coding agents help me move faster on my own projects. I review every line, keep secrets out of every prompt, and run privacy checks in CI on every push.

timeline
    title Career roadmap
    section Foundations
        Earned : Google Cybersecurity Professional Certificate
               : APU Professional Certificate in Computer Systems and Networks
               : CompTIA Security+ Certificate
               : ISC2 Certified in Cybersecurity
        Founded : CTRL+J Juice Lounge & IT Hub LLC
    section Now (2026)
        Working : Computer Technical Assistant, Fonda-Fultonville CSD
        Shipped : AI Guardian Web Shield on the Chrome Web Store
                : Cole Learning Labs with 100 free K-12 activities
        Studying : Cybersecurity senior at UAlbany
    section 2027
        May : B.S. in Cybersecurity, University at Albany
        Next : Begin CTE certification to teach Computer Technology
    section Beyond
        Career : K-12 security operations and identity engineering
               : Human-centered security products and accountable AI
        Community : Reopen CTRL+J as a community technology space

I'm building toward work in:

• K–12 technology education through CTE certification, so students meet cybersecurity, networking, and responsible AI before they meet their first phishing email.
• Security Operations and threat detection, where plain-language explanations help teams and users act faster.
• Identity and Access Management and Zero Trust engineering for schools and public institutions.
• Human-centered security products that respect privacy, accessibility, and user choice.
• Responsible and accountable AI systems that keep humans answerable for what machines decide.

My long-term goal is to bridge technical security, ethical governance, accessibility, and real-world human needs.

• B.S. in Cybersecurity, University at Albany
• Google Cybersecurity Professional Certificate
• Professional Certificate in Computer Systems and Networks
• CompTIA Security+ Certificate
• ISC2 Certified in Cybersecurity
• TryHackMe learning paths: Cyber Security 101 and Pre Security
• Ongoing practice through TryHackMe labs, local vulnerability audits, defensive simulations, and open-source project development

• Security should be technically strong, ethically grounded, and understandable to the people it protects.
• AI must not become a place where human decision-makers hide responsibility.
• Privacy, dignity, accessibility, accountability, and meaningful user choice belong in the design from day one.
• Defensive security should protect and contain, not retaliate.
• The best security lesson is the one a student can explain back to you.

• 🌐 Cole Learning Labs
• 💼 LinkedIn
• 🐙 GitHub
• 🛡️ AI Guardian Web Shield on the Chrome Web Store

_{This profile embeds no third-party images or trackers. See security reporting and repository privacy.}