Fractional CTO.

Architecture is a leadership decision. Most early-stage tech companies don’t fail because the code was wrong. They fail because the sequencing was wrong, the trade-offs were never surfaced, and the team was hired before there was a thesis to hire against. A fractional CTO is the senior engineering operator who shows up part-time, owns those decisions with the founder, and leaves behind a team that can outlast them. I’m the second engineer in the room when founders are deciding what to build, what to throw away, and who to hire.

Five situations bring most founders to the table.

Pre-PMF founder. Technical or not, the work is the same. Figure out what to build, ship the smallest version that tests the thesis, know when to stop refining and start selling. The hardest call here is when to stop. I make it explicit.

Non-technical founder. You have a market thesis (or you’re still finding it) and no technical co-founder. I make the architecture, hiring, and build-vs-buy calls you don’t have the context to make alone, sharpen the positioning where it’s still soft, and translate the engineering reality back into the business decisions you do own.

Fundraising. Six to eight weeks before a round. We frame the technical and GTM narrative, fix the structural gaps, and rehearse the questions a serious lead will ask in DD. The diligence pack ships before the investor requests it.

Investor side. A portfolio company needs a technical read, or you’re sizing up a new deal. I run the diligence and translate findings into deal terms. See /tech-due-diligence.

Post-rewrite or post-pivot. A major architectural or positioning change just landed. Someone outside the team should pressure-test the assumptions before the burn rate makes them irreversible. Better now than at the next board meeting.

When it doesn’t: you want a strategy deck without execution; you need a full-time CTO running a 20-person org; you want pure staff augmentation; the engagement is “manage this team for me, send me weekly reports.” Different role.

I’m the second engineer in the room when founders are deciding what to build, what to throw away, and who to hire. The first engineer is usually the founder or the lead, holding the system in their head. My job is not to replace them. My job is to give them a peer in the room, someone who’s seen this decision shape ten companies and can name the trade-off out loud before it gets quietly made. The decisions sit with the founder. The articulation, the alternative, and the second look come from me.

Most early technical mistakes are not code mistakes. They’re decision mistakes made too late. Wrong stack chosen six weeks before there was a customer. Senior engineer hired six months before there was an org to lead them. Migration deferred until the day it had to ship. The audit fixes none of this. The leadership decision, surfaced early, fixes most of it. Most of my work is timing.

My job is to reduce irreversible choices, keep momentum, and leave behind a team that can outlast me. The reversible decisions get made fast. The irreversible ones (data model, hiring, regulatory posture, technical debt that compounds quietly) get a meeting and a written record. By month six, the team should be making those calls without me. If I’m still in every conversation, I haven’t done the job.

The work isn’t only code. It’s also positioning, sales decks, cold outreach, and the early fundraising story. Most fractional CTOs don’t do that. I do because at pre-PMF, the technical roadmap and the market story are the same conversation. When I positioned e-invoice.be for its expansion phase, the obvious frame was the EU mandate clock: Belgium live since January, Germany mid-rollout, France landing in September. I refused it. Mandates are a deadline, not a moat. The opportunity sits one step beyond compliance, in what mid-market finance teams do with structured invoice data once they have it. The position became a dual moat: developer-grade Peppol infrastructure for teams that want to build, plus a no-migration PDF-via-email path for the larger group locked into legacy ERPs. The technical roadmap followed the position, not the other way around.

Operationally: I work embedded as a co-builder. Eight hours a week is typical for the retainer. Diagnostic engagements (tech audit, product sprint) run one to two weeks. Three to six months is the usual horizon for the retainer. Remote-first across Europe, with occasional on-sites in Belgium and the Netherlands. The first two weeks are always a paid diagnostic, written down, before either side commits to anything longer. Full breakdown on /engage.

Pricing is fixed-fee or fixed-day-rate, never time-and-materials. Engagements run from €2,000/day for advisory, through €8,000 to €20,000 for one-to-two-week sprints and diagnostics (tech audit, product sprint, product marketing sprint), up to €8,000/month for the embedded co-builder retainer. Full breakdown on /engage.

Full-time CTO. Right when you have an organization to run and €200K+ a year of CTO budget. Wrong pre-PMF. The full-time hire builds a team and a process, both of which are overhead until there’s enough traffic to justify them. If you’re at five engineers and €5M ARR, hire one. If you’re not, you’re paying for capacity you don’t yet need.

Technical co-founder. The right call when you’ve found the right person. Looking for one is a 6 to 12 month search and the failure rate is high. Most founders don’t have 6 to 12 months. The honest fractional engagement is “I do this with you for nine months, and we either find your co-founder during it or we don’t, and either way you ship.”

Consultant. A consultant produces a deck and leaves. The work product is a recommendation. A fractional CTO writes the code, makes the hire, ships the product, and stays close enough to the work to know whether the recommendation actually held. Different work product. Pick based on whether the answer to your problem is a slide or a build.