How we work together
Three ways to start, in ascending commitment. Every engagement ends with you owning the entire system — the vault, the AI configuration, the documentation, and the training materials.
€500 founding-client rate — €750 thereafter
- 90-minute call: intake + screen-share walkthrough of your current setup
- Written diagnostic, 6–10 pages (PDF)
- 15-min video walkthrough of where your project memory is leaking
- 30-day fix plan you can implement yourself — or hire me to install
€4,500 founding-client rate — €7,000 thereafter
- Document audit + ingestion plan
- Vault structure tailored to your workflow
- Ingest of up to 3 GB of existing project documents
- Wikilink concept index — people, projects, vendors, decisions
- AI query layer configured against your vault
- 2 training sessions with your team
- 30 days of post-launch support
from €2,000/month · offered after a successful Build
- Weekly ingest of new documents
- New concept / people additions
- AI prompt tuning as the project evolves
- 1 monthly review call with the principal
- Dedicated channel for ad-hoc questions
Why the rates are this low — read this
I’m taking on the first two paying clients at the reduced founding-client rates above. In exchange, you give me permission to publish the engagement as a real, named case study when it’s done. I keep all data confidential during the engagement; the case study goes out only when you’ve reviewed and approved every word.
After those two cases close, the rates go to standard pricing (€750 audit, €7,000 build) and case studies become opt-in only.
Why I’m doing it this way: the methodology is real and the engine is in production — but the named-client case studies don’t exist yet. The fastest honest way to get there is to do excellent work for two early clients at a price that makes the bet easy for them.
If you’d rather pay standard rates and skip the case-study clause, that option is also on the table — just say so on the call.
FAQ
Have you done this for a real AEC client yet?
The Project Memory engine is in production daily — on my own context, with 7,571 lines of Python, 12 skills, 11 vaults. Green Score ran end-to-end on a real Brazilian residence model. The named-client AEC engagements are what I’m building right now with the two founding clients — that’s exactly why the rates are reduced.
What about my data — privacy, confidentiality, EU AI Act?
The deliverable is fully local: vault on disk, no cloud sync of curated content. LLM calls only against configured zero-data-retention agreements. The system is built aligned with EU AI Act art. 4 (AI literacy) and art. 50 (transparency), with documented retention and redaction patterns.
What if my projects aren’t in Portugal?
The Portugal positioning is where the wedge is sharpest, because of regulatory fluency. The technical engine is fully transferable to any AEC market. If you’re outside Portugal and the rest of the pitch resonates, book the call and we’ll sort it out.
Why not just use ChatGPT with my files?
Three reasons. One — your data ends up on someone else’s training set unless you’ve negotiated a zero-data-retention agreement. Two — ChatGPT doesn’t understand the conceptual graph of your project (people, decisions, vendors, permits, drawings), so it gives shallow answers to deep questions. Three — there’s no audit trail, which means it can’t satisfy licensing, insurance, or dispute requirements.
How long does the engagement actually take?
Audit: 90-minute call + 5 business days — your diagnostic within a week. Project Brain Build: 3–4 weeks from kickoff to handoff. Retainer: ongoing, monthly cadence.
What happens after the engagement ends?
You keep the entire vault, the AI configuration, the documentation, and the training materials. If you renew on retainer, I keep the system fresh; if you don’t, it still works — you just maintain it yourself.
One 15-minute call
No pitch deck. No script. I diagnose your specific situation on the call. If it’s not a fit, I’ll tell you on the call.
Book the call