Enterprise IT is rarely just “the thing is broken.” Usually, the real work is figuring out what actually failed, who it affects, what changed, and how to make sure the same issue does not come back wearing a different outfit.
I work across enterprise IT operations, endpoint deployments, Microsoft 365, identity, VPN access, documentation, vendor escalation, and AI-assisted troubleshooting, with a focus on making complicated systems clearer, safer, and less annoying for the people who actually use them.
// Bubbly delivery. Forensic receipts. Systems fixed properly.
// What I Fix
I sit in the space where the technical issue, the user panic, the vendor excuse, and the business deadline all arrive at the same time. Lovely little disaster bouquet.
// AI Reliability
My AI work comes from a practical IT problem: faster answers are only helpful when they are also grounded, testable, and safe to use.
I treat AI like an operational system, not magic. It has hidden instruction layers, retrieval boundaries, source risks, tool limitations, and a very annoying habit of making guesses sound polished. My work is about building the practical guardrails around that: source-grounded workflows, clear assumptions, rollback language, and human judgment staying firmly in the loop.
my AI reliability stack
Debugging the interface between probabilistic text generation and workplace accountability.
// Stack Diagnostics
The problem is not always the laptop, the user, the network, or the app. Sometimes it is the handoff between the browser, the auth layer, the API, the database, the vendor system, and one tiny setting quietly choosing violence.
I am not pretending every system is mine to rewrite. I am very good at reading where a web-app problem is likely breaking: front-end behaviour, back-end logic, browser/session weirdness, permissions, vendor boundaries, or the lovely little swamp where all of those overlap.
A lot of technical support is really translation between what the user sees and what the system is doing underneath. I can trace the shape of a failure through login flow, token behaviour, permission scope, API response, database state, and handoff logic without getting distracted by the loudest symptom.
I like evidence packets. Browser errors, stack traces, timestamps, screenshots, logs, reproduction steps, version context, and Git history all help separate the actual break from the dramatic fog machine around it.
When the fix belongs to a vendor or dev team, the escalation still needs to be clean. I package the symptom, environment, user impact, tested paths, failed assumptions, and likely failure layer so the issue is harder to bounce back with a decorative shrug.
// Proof
A few examples of the actual operating surface.
Organized and led the physical swap, configuration, and deployment of over 400 Windows endpoints ahead of a strict October deadline tied to Windows 11 hardware compatibility.
This included coordinating physical logistics, workstation setup, standard application installation, user profile continuity, Outlook/profile setup, mapped drives, printers, peripherals, and location-specific requirements across multiple business units.
Supported rollout planning, documentation, scheduling, communication, and user guidance for MFA across a 700+ account environment.
The work required technical accuracy and user adoption at the same time: clear instructions, realistic support expectations, and enough structure to reduce confusion during the transition.
Managed Active Directory permissions and access structure for a FortiClient VPN deployment supporting an under-100-user off-site footprint.
The goal was secure remote access that matched real operational need: controlled, explainable, reversible, and tied to the correct users and groups.
Troubleshoot application and vendor-system issues by separating user symptoms from account state, access scope, device behaviour, network path, backend/admin configuration, and third-party platform boundaries.
This is the SaaS support shape: identify which layer is actually failing, preserve the evidence, document what changed, and package the issue clearly enough for users, vendors, developers, or leadership to act without turning the ticket into a fog machine.
Support identity and access issues across Microsoft 365, Exchange, Active Directory, MFA, VPN access, mailbox/profile resources, and user-facing authentication workflows.
This gives me the practical pattern recognition for SaaS auth issues: separate the user’s visible failure from the actual access layer, credential state, permission scope, session behaviour, or admin-side configuration before escalating or changing anything.
Built and used AI workflows for troubleshooting, research, source-checking, documentation drafting, and diagnostic logic testing.
The goal is not to let AI make the decision. The goal is to make the research path faster, the source trail clearer, and the assumptions easier to inspect before judgment is applied.
Package symptoms, logs, system states, timelines, reproduction details, and failure patterns so third-party vendors have less room to bounce unresolved issues back onto local hardware.
Good escalation is not “it is broken.” Good escalation is a clean evidence packet showing what failed, what changed, what was tested, which layer is likely involved, and what outcome is needed.
Support enterprise voice and communication-routing workflows across multiple locations, including call queues, IVR logic, forwarding rules, and user/extension configuration.
This work keeps business communication moving reliably by making sure calls route to the right people, departments, and escalation paths without dropped handoffs, dead ends, or avoidable confusion.
// How I Work
Technical translation is the layer between a correct fix and a usable fix. The fix changes depending on who needs to act on it.
Plain-language support
Users do not need a dissertation on the failure mode. They need to know what happened, what to do next, what to expect, and whether they can keep working.
I turn technical mess into instructions people can actually follow, so the fix does not depend on the user already speaking fluent IT gremlin.
I translate the failure into clear next steps: what happened, what to expect, and if they can keep working.
Provide low-friction instructions and exact support paths so they don't get lost in the transition.
Set clear boundaries on what they can trust the AI to do, and when to bring in a human.
I gather exact symptoms from the user without making them learn IT jargon, keeping them out of the support loop.
// Resume
The short version, for people who need the receipts quickly.
profile
Dynamic systems generalist with a keen eye for detail and a strong drive for operational efficiency. Known for building strong working relationships, learning quickly, and turning complex technical problems into stable, repeatable fixes. Experienced across multi-location, multi-vendor IT environments, with a focus on root-cause troubleshooting, clear communication, and practical improvements that keep systems working properly.
current
Budds’ Group of Companies
Support centralized IT operations across a 700+ user, multi-location enterprise environment, with work spanning endpoint lifecycle, Microsoft 365, Exchange, Active Directory, MFA, FortiClient VPN, remote access, documentation, vendor escalation, specialized business systems, and AI-assisted troubleshooting.
previous
Geek Squad with Best Buy
2022 – 2025
Canada Computers
2021 – 2022
UKFast
2015 – 2018
education
Richmond American University London
BA (Hons) Social Sciences, major in Psychology with minors in Philosophy and Math
Bishop’s College School
CESI accredited secondary diploma. AP courses: Calculus, Physics, Biology, English.
outside the ticket queue
Outside of work, I build and optimize custom PCs, tinker with smart home automation, and treat technology as something that can always be improved with enough persistence. I am also a gamer, pianist, clarinetist, runner, golfer, tennis player, and dog person.
// Contact
For recruiting, hiring, AI implementation, IT operations, technical customer success, or roles where systems need to be made clearer and less annoying, contact me directly.