A cascade of vulnerabilities: security cannot wait
October 2026 alerts span runtimes, dependencies and operating systems. Keeping up requires named owners, time to act and evidence that fixes reach the running service.
The Diagnosis
The roadmap is slipping, but leadership and the team disagree on why? I look at the code, organisation and product decisions together. Choosing a remedy before understanding the problem often means paying twice.

Action plan delivered within 7 days.
When to use it
This assessment examines the problems before you hire, rebuild the product, speed up delivery or commit to a longer engagement.
The roadmap slips but nobody agrees on the cause
I separate tech debt, product focus, team dependencies and implicit leadership tradeoffs.
The lead developer or internal CTO is carrying too much
We clarify what belongs to leadership, organization, documentation or actual team capacity.
You are preparing a major decision
CTO hiring, fundraising, acquisition, major client, certification, rebuild or architecture change.
What I look at

Architecture & tech debt
Areas slowing the roadmap, increasing risk or concentrating too much knowledge.
Delivery & quality
How work moves from idea to production: CI/CD, tests, incidents, releases and ownership.
Team & governance
Roles, responsibilities, rituals, decisions and communication between leadership, product and engineering.
5 days
Framing
A call to understand context, decisions to make and key people involved.
Review with the team
Focused interviews, flows, architecture, backlog, incidents, available data and team rituals.
Readout
Leadership summary + prioritized action plan, with risks and tradeoffs made explicit.
Deliverables
Risk map
What can break, slow down, cost money or block the next business step.
30 / 60 / 90-day priorities
What to start now, what to prepare, and what not to do yet.
Recommended engagement format
A focused project, a fractional CTO, team coaching, hiring or a plan your team can carry out internally.
Is this a classic technical audit?
The diagnosis focuses on decisions involving technology: product, organisation and business priorities. I review code and architecture where useful. A comprehensive code audit would need a separate scope.
Do you need access to the whole codebase?
Not necessarily. We decide the access level together. Often interviews, diagrams, backlog, incidents and a few key areas are enough for a reliable first read.
Does it also audit the tech team?
Yes, on the organization side: roles, responsibilities, rituals, decisions and communication between leadership, product and the tech team, based on targeted interviews. I read them alongside architecture, technical debt and delivery: the same situation often spans several areas.
What happens afterwards?
You choose what comes next: carry out the plan internally, start a focused project or continue with a fractional CTO.
We start from your real situation and check whether the diagnosis is the right first step.
October 2026 alerts span runtimes, dependencies and operating systems. Keeping up requires named owners, time to act and evidence that fixes reach the running service.
A defaced site, backdoors and a hidden miner: how the investigation into a WordPress staging site led to rebuilding an entire machine.
Two sale processes, each requiring a month of preparation: the key people, licences, contracts and risks to document for the buyer’s review.
€550 a day excluding VAT, a €3,000 diagnostic: budgets by time commitment, available public benchmarks and the differences from a salaried hire.
You are already paying for your technical debt, every week, on the team’s payroll line. Six cost items, their formulas, a fictional six-developer example, and what I would do with the number.
Reviews, approvals, release slots and concentrated knowledge: examining a team’s waiting time, the effects of hiring and the DORA metrics.
Organising a team, bringing in expertise or preparing a product: three situations for defining a freelance CTO’s decisions, follow-up and time commitment.
An invoice with the wrong address, three hundred avoidable refunds, a domain for sale at 50 dollars. Five cases covering code, organisation, autonomy, tooling and security.
One order for Lyon, one invoice for Paris, a single address field. What a “small field” reveals about software nobody dares to touch any more, and four test cases for preparing the change.
Three hours to fix and test, six days to be allowed to release. The ticket is done, support is still refunding. In this fictional case, the wait adds ten hours of support work.
120 paid orders, 118 invoices sent, one green light. A fictional case separating website availability, sending checks and recovery of failed invoices.
A forgotten Brevo script in a website, an expired domain still trusted, hundreds of applications someone once said yes to. Three checks for establishing how these integrations are used and what they can access.
Hundreds of Slack exceptions every morning, a team split in two, a priority client waiting. Then a month without a single new feature, negotiated with the CEO. I still stand by it: in my view, the most profitable decision of the year.