AGX Consulting is a solo architecture-review practice. I partner with engineering teams to pressure-test systems, surface latent risk, and produce decisions you can act on inside a single sprint — not a 90-page deck.
AGX Advisory is built around one discipline — architecture. I work with engineering teams to pressure-test existing systems and design new ones from the ground up — uncovering hidden risk, making critical decisions, and delivering recommendations clear enough to act on immediately.
Every engagement is run personally, end to end. No subcontracting, no juniors, no deck-driven deliverables. You get a written report — prioritised, defensible, and short enough to actually read.
Every engagement ends in something written — a blueprint, a memo, a register of findings. No slideware, no recommendation theatre. The right format depends on where you are.
Full end-to-end architecture design for new platforms — from concept to blueprint. Ideal for founders and CTOs starting a complex build who want the system designed properly before a line of production code is written.
Assessment of an existing system — identifying risks, gaps, scalability ceilings, and technical debt. A great first engagement that often leads to bigger work once the picture is clear on paper.
Independent technical assessment for investors or acquirers evaluating a technology platform before committing capital. Clear-eyed read of the system, the team, and what you'd actually be buying.
Ongoing access to senior architecture advisory on a fixed monthly basis. Covers reviews, guidance, team Q&A, and strategic input as the product evolves — a second pair of eyes always on standby.
A structured engagement to assess and improve how your engineering team works — covering delivery workflows, sprint cadence, code review practices, team structure, and development standards. The outcome is a leaner, faster, more consistent engineering operation.
Every engagement runs the same way: a short call to see if it's the right fit, a written scope you can take or leave, then the review itself — on a calendar, with a written deliverable at the end.
A short, confidential call to understand the system, the stakeholders, and the question worth asking. No obligation — you leave with a one-page memo either way.
The review itself. Code, infra, data, runbooks, the people on call — read carefully, in context, against an agreed brief. One written report at the end. No deckware.
After the report, an optional standing advisory — a second opinion on designs, ADRs, and incidents while you act on the findings. Bi-weekly, written, async-first.
The fastest way in is a 30-minute scoping call. You describe the system and the worry; I describe how (and whether) I would review it. If we proceed, you have a written scope inside a week.