Handled repeated storage and file platform migrations across — classic Windows file servers into NAS appliances, NAS and file servers into SharePoint, Google Drive into SharePoint, SharePoint back into Google where the fit made more sense, and Dropbox estates consolidated into either Microsoft 365 or Google Workspace.
On paper these projects are often sold as “just move the files.” In reality, the hard part lives in permissions, path depth, broken naming habits, sync behaviour, stale shares nobody documented, users who still think the X: drive is part of the laws of physics, and making sure the new platform does not just hold the data but actually improves how people work with it.
The goal was never only to copy data from point A to point B. It was to land the business in a cleaner, more supportable place — access mapped properly, sync understood, collaboration improved, legacy clutter reduced, and the new platform shaped around real operations instead of just becoming a shinier mess.

Delivered repeated mailbox and collaboration migrations — Google Workspace to Microsoft 365, Microsoft 365 back to Google, legacy IMAP and POP platforms into both, and the one people consistently underestimate: Microsoft 365 tenant-to-tenant cutovers.
These projects are never just about moving mail. The real work lives in identity cleanup, domain cutovers, alias conflicts, Outlook rebuilds, mobile reconfiguration, delegated access, shared mailboxes, calendars, DNS timing, and all the small edge cases that can quietly turn a weekend migration into a Monday morning disaster.
The objective was always the same: get users cleanly to the other side with mail flow working, authentication aligned, devices reconnected, shared access restored, and the business moving forward with as little disruption as possible — whether the driver was licensing, mergers, restructuring, or simply a better platform fit.

Walked into a 3-building hotel running a completely flat network — no VLANs, no documentation, dedicated cables and switches for every "network," and APs wired however someone felt like it that day.
Step one: figure out what was actually there. Traced every cable, mapped every switch port, and built the logical diagram from scratch.
Step two: design the proper architecture. Converted the entire network to VLANs, aggregated uplinks to create bonded redundant paths across the buildings, and re-tagged every switch port and AP. Divided and locked down inter-network permissions so guests, staff, and systems only see what they're supposed to.
Result: a network that could actually be managed, documented, and scaled.

Two-building church campus where the network's brain lived in a Windows Server — doing DHCP, VPN, and routing living in a switch, all the firewall duties it had no business doing. VLANs existed but were a mess: overlapping, unused segments everywhere, and access between networks wider than it needed to be. Audited and documented the full topology across both buildings. Migrated DHCP, and VPN from the Windows Server into the firewall, migrated also the routing to the firwewll, where it belonged. Cleaned up and consolidated the VLAN scheme, killed off unused networks, and tightened inter-VLAN access rules so each segment only reached what it actually needed.

Built a high-redundancy private cloud for an MSP that wanted the capability of Microsoft 365, Google Workspace, and modern hosting platforms — without handing over control of the whole estate to third-party vendors.
Started with the core: resilient compute, Ceph distributed storage, and the plumbing to survive node or service failures without turning every outage into a business event. From there, layered the real-world stack on top: email and collaboration, identity and SSO, password management, device management, remote access, monitoring, and SIEM.
The end result was not just self-hosted for the sake of it. It was a platform with data sovereignty, predictable costs, no per-user lock-in pressure, and the freedom to tune the stack around operational reality instead of vendor limitations.

Corporate DNS migration across two redundant datacentres serving roughly 250,000 daily active users — the kind of project where you only get noticed if it goes badly.
My role sat in the networking and security layer: test it hard, break it in the lab before production can break it for real, and make sure failover, performance, and visibility were solid before cutover day.
Once live, monitoring proved its value fast. It exposed a customer blasting over a million DNS queries a day, which let us intervene early, work with the customer, and fix the issue before it became a platform-wide problem.

Consolidated multiple mail systems into a single platform serving more than 50,000 accounts — not just a migration job, but a stability and reputation recovery project.
A major pain point was outbound spam. Inbound filtering existed, but outbound control was weak, so reputation incidents kept coming back. I build and tune the protection properly on both directions, reducing recurring spam events and giving the platform a much cleaner operational baseline.
To protect deliverability, I also split a /23 public IP pool into relay groups based on domain size and sending patterns. That made it possible to isolate noisy senders, recover blacklisted IPs faster, and stop one bad actor from dragging down everyone else.

Built and hardened a shared hosting platform on cPanel and CloudLinux at a time when “just migrate it” would have broken too many customer sites.
The challenge was not spinning up the server — it was moving real production workloads with different PHP requirements, old applications, and all the usual customer surprises that appear halfway through a migration window.
Using PHP selector isolation and careful staging, I migrated sites with minimal breakage and turned a scattered environment into a cleaner, more secure multi-tenant platform that could actually be operated with confidence.

Part of the team that helped turn a traditional ISP environment into a more cloud-ready service platform for customers.
My side of the work lived in networking and security: hardening the perimeter, supporting the consolidation of hosting and mail infrastructure, and helping create the operational foundation the customer-facing services would rely on.
A lot of the later wins — DNS, hosting, mail, and service stability — came from this groundwork. It was the base layer that made the higher-profile projects possible.

Ran central management and visibility for more than 250 Fortinet devices across the estate, bringing logs, policy visibility, and operational control into one place instead of living blind across dozens of isolated boxes.
That role also tied directly into vulnerability management and ISO 27001 work, where security could not stay theoretical — it had to be evidenced, repeatable, and support audit requirements without slowing the business to a crawl.
When things went sideways, I handled escalations with Fortinet TAC directly, bridging the gap between vendor support and real production pressure until problems were resolved properly.