Brandon Moore.

• Backend systems, language-agnostic. The work I do is architecture, multi-tenant SaaS patterns, integration plumbing, and queue-and-event design. The language is whatever the operator’s existing stack uses. I’ve shipped production work across multiple backend languages and am comfortable picking up the one a given client’s codebase already runs on, rather than forcing my preferences into someone else’s environment.
• Self-owned, bare-metal infrastructure. I run my own production server fleet on hardware I own — no third-party PaaS sitting between me and a client’s critical systems. When a hosted vendor goes down, my pipeline keeps moving. When a SaaS pricing model changes, my margin doesn’t. For operators whose distribution depends on always-on infrastructure, that independence translates directly to the work: I can recommend self-hosted paths because I run them myself, not because I read about them.
• Any client environment, hosted or on-prem. I work inside the operator’s existing stack as the default — their cloud, their on-prem, their datacenter, their domain controller. The bare-metal independence is mine; the deployment shape is theirs. I don’t arrive insisting an operator migrate to my infrastructure to get the work done.
• ERP and pipeline integration. Through documented APIs when they exist, through other paths when they don’t. Browser-resident augmentation, direct database integration, queue-and-event architectures, custom protocol bridges — the right one is whatever closes the gap fastest without breaking the operator’s existing tooling.
• Label and printer protocols. ZPL, Zipher, EPL, raw TCP. Direct-to-printer pipelines when middleware is the bottleneck. Per-station queue architecture when routing is the failure mode.
• Networking and infrastructure plumbing. VPN-to-on-premise paths, certificate handling, subnet diagnostics, DNS, BIND, firewall and NAT rules — the kind of work that comes up when an “obvious” hardware fault turns out to be a routing or auth problem nobody thought to check.
• Hardware and mechanical capability. 3D modeling and 3D printing for one-off mechanical fixes when the gap is physical. A tool I reach for only when the right answer is mechanical, not when the right answer is software.
• Diagnostics on the floor. The part that makes the rest matter. Walking the line, watching the handoffs, finding the gap nobody upstream could see because the gap was between two things they each owned.

Each engagement uses a different subset. The fit is what matters, not the inventory.

Once you have walked a distribution floor and watched the COA, the label, the printer, and Metrc fail to agree with each other on the same package — and watched the vendors involved each correctly insist they are not the problem — you cannot un-see it. The gap is real. It is not getting smaller. The DCC’s recall data confirms what the floor confirmed: nearly half of California cannabis recalls in the last twenty-eight months are label or cannabinoid-pipeline failures, and the rate is rising.

I have chosen, deliberately, to work only on regulator-aligned distribution compliance. Not gray-market. Not plant-touching. Not speculative. The work is grounded in public regulatory data and verifiable operational outcomes. That choice is structural to the practice, not a marketing posture.