Chief Technology Officer, platform builder, and applied AI product engineer
A working history of cloud platforms, applied AI systems, developer tooling, and product surfaces that still make sense after launch.
Start with the offerBuilding infrastructure, products, and operating paths for high-context enterprise software. The useful system is the one a team can run twice, explain under pressure, and improve without losing context.
Where I am useful.
I am most useful when the work has enough context that a generic implementation plan starts lying: Archibus upgrades, managed workspaces, applied AI support paths, platform rescue, and developer tools that need to survive real users.
Upgrade readiness, repeatable test/UAT environments, performance triage, and support paths around real WebCentral constraints.
Agent workflows, context boundaries, token-cost discipline, and human review loops for teams that need answers they can defend.
Cloud workspace platforms, deployment visibility, auth boundaries, billing surfaces, and operational feedback loops.
Editors, CLIs, dashboards, worktree workflows, and automation that reduce handoff friction instead of adding another ceremony.
Focused dossiers.
Archibot turns high-context WebCentral support work into repeatable environments, safer diagnostics, and AI-assisted tooling that stays close to the source material.
Open fileThe workspace work is about turning fragile local rebuilds into productized environments with identity, templates, billing, support paths, and visible operating evidence.
Open fileDexaga was the product path from consulting knowledge to a tool: identify available data, query it, aggregate it, and make related operational context visible.
Open fileSteamPanno was a reminder that the delivery shape is part of the product. A one-time creative tool should not ask for more trust than the output deserves.
Open fileThe useful agent system is less like a crowd of autonomous workers and more like a disciplined sidecar workflow with clear ownership, evidence, and final review.
Open filePublic work record.
This is the current chapter: taking high-context facilities management software work and turning it into systems teams can actually run, support, and improve.
Dexaga helps users identify, query, aggregate, and understand data already available in operational databases without needing to know SQL.
This period connects the Counterpointe R&D work to Dexaga: taking messy operational systems, building tools around them, and making the result usable by people who are not sitting inside the code.
This was the period where I learned to stop accepting slow enterprise software as inevitable and started building tools that changed the shape of the work.
This is where the Archibus and facilities management software domain became the base layer for the rest of the work.
This was lower-level systems work: hardware support, test routines, simulators, version-control discipline, and Unix-like operating environments.
This was early mobile work with the whole release path attached: application code, database work, security concerns, signing, and publishing.
Howard University and the systems work around it.
Computer Science, with the useful parts happening both in class and in the lab.
Managed a student computer lab with 50+ machines and helped keep real users working.
Built cross-platform VM networks, Xen Ubuntu hypervisor labs, and bridged web, app, and LDAP services.
Worked on firewall upgrade ideas, intrusion prevention, and a GPS-based lane-departure warning system.
Dense inventory from the work.
KeyControl had to leave my memorized WebCentral setup and still install, run, and pass checks for someone else.
Product revamp Read issueThe container is the easy part. Real Archibus workspaces also need databases, licenses, restore paths, startup scripts, IDE access, diagnostics, and support evidence.
Managed workspaces Read issueArchibot is a repeatable Archibus lab with AI attached. That means the token math has to be honest from the start.
Archibot economics Read issueParallel agents are useful when the work is actually parallel. They do not remove the serial review, merge, and judgment path.
Agent systems Read issueISM is not a generic software shop. Most of the work lives in the hard middle around Archibus, operations, data, support, and tools that make the work repeatable.
ISM Services Read issueMinIO's community edition had been heading toward maintenance mode for months. I noticed when Reddit made it embarrassing to miss.
Storage migration Read issueFlat-rate access to frontier AI coding tools was always a subsidy. Coding agents made the cost too visible to keep hiding.
AI economics Read issueThe research cluster proved Archibus could run in Kubernetes. The next problem was that too much of it still required me.
DevOps archaeology Read issueA hosting question, Authentik, and a pile of manual Kubernetes scripts. The first real lesson was not that self-hosting is cheap.
DevOps archaeology Read issueThe first useful Archibot knowledge-system lesson was not a vector database. It was learning how to make private software knowledge retrievable at all.
RAG archaeology Read issueA Steam library poster exposed a larger product question: how much trust should a one-time tool ask for?
Experiment Read issueAvailable for the right platform, AI, and product engineering conversations.