Independent technical oversight so that leadership regains control, reduces tech debt and makes decisions without depending on a single voice.
Three questions. No form, no cookies. No strings attached.
Critical software, no independent oversight, and a concrete reason. That's exactly the kind of conversation that makes sense.
Based on what you've told me, my profile isn't what you need right now. But if you want to validate it on a quick call, no problem — if I don't see a real fit, I'll tell you directly.
Three symptoms any board member or GM recognises immediately. If two of them are true, we already have something to talk about.
They've been delivering software I don't fully understand and that gets more expensive to maintain every year. When something breaks, I have nobody with the judgement to defend my position.
The system runs a meaningful part of daily operations, it has been growing for years, and deployments are frightening. I can't tell whether the issue is the team, the process or the code itself.
Copilot, Cursor, Claude — they're inside the development flow with no governance, no real productivity metrics and no security policy. Technical debt is silently accelerating.
You don't have to guess which engagement fits. The situation decides.
Independent technical judgement month after month. No conflict of interest, no analyst rotation.
View full profile → 02Full picture of your most critical system, translated into risk and business cost.
View full profile → 03Phased work on production code. Without stopping the service, transferring judgement to the team.
View full profile → 04Turns your team's AI adoption into a measurable advantage, not invisible debt.
View full profile → 05Second opinion before signing. Debt translated to euros, roadmap validated against the real team.
View full profile →Auditable metrics on high-availability critical systems. Full dossier available after a first conversation.
Critical processing platform with high operational demand. Redesign towards asynchronous distributed architecture with zero downtime and without rewriting the core engine.
Critical transactional platform handling high volumes of concurrent operations. Data model redesign and adoption of event-driven streaming, drastically reducing response times and infrastructure pressure.
Critical corporate component spanning multiple technology teams. Architectural transformation to Hexagonal Architecture, testing framework built from scratch, CI/CD pipeline standardisation and governed delivery process.
Highly complex inherited codebase with a one-month delivery window. Applied a proprietary four-flow canonical AI playbook: incremental refactor, AI-assisted test generation, adversarial PR review and automated documentation.
If your company relies on critical software and needs an independent second opinion on what is being delivered, we probably fit. The right column describes the three scenarios where I have nothing useful to offer.
Regulated environments, high-availability systems, data. Founder experience early on. Tech Lead of a team I left fully autonomous. Today, independent technical supervision for companies that need a second opinion on what their vendor — or their own team — is actually delivering.
I spent nine years rescuing critical systems in environments where software failures are not an option: regulated banking, cybersecurity, data engineering. The pattern was always the same — someone had trusted a single voice for too long to evaluate what was being delivered. Today, that independent voice is mine. What in construction would be called an independent clerk of works: the technical expert on the client's side, not the vendor's.
I led the stabilisation of a critical communication component serving multiple internal teams — handed over without tests, without documentation and without governance. Before that I did the equivalent on a high-volume processing system. That is what gives me the judgement to oversee yours.
Explicit filters. If we fit, we fit. If we don't, we save both sides a long conversation.
The audit covers code, architecture, infrastructure and resilience — not regulatory accreditations. DORA, NIS2 or ENS require certified bodies; if you need it, I'll route you to the right profile.
Not my turf. For pen-testing or judicial expert work I recommend specialists with offensive certification. If you tell me the context I can suggest the right profile.
No. The model is supervision and directed modernisation over existing systems, not software delivery. If you need greenfield I'll route you to a trusted development boutique.
No — outside my verifiable expertise. I decline the engagement even at a premium: better to be clear upfront than to compromise the quality of the work.
No. The model is fixed scope or monthly retainer with SLA. Hourly billing creates the wrong incentive: I would earn more if the project runs long; you want it done well and fast. Fixed price aligns us both.
Remote by default, with occasional visits to Madrid or Andalusia when they add real value. On ongoing supervision, communication is typically asynchronous with a guaranteed 48-hour business response.
Yes. Standard NDA before kickoff. Published cases are anonymised with explicit client consent. IP of the executive report transfers after full payment.
Boutiques sell teams and projects. This is the founder's judgement with limited, publicly visible capacity and an SLA. I don't replace your team: I work alongside it and leave it more capable.
No sales deck, no pitch. Tell me what you're dealing with and I'll tell you honestly whether we're a fit and how I can help.
Based on the assessment, your situation fits. Fill in what you consider relevant and I'll reply within 24 hours.
You want to validate it on a call. No problem — if I don't see a real fit, I'll tell you directly.