Discovery
Users, workflow, constraints, risks, and the smallest release worth funding.
Custom software · Portugal and Europe
I design and build web products, internal tools, integrations, and data systems for teams with a specific operational need. The first job is to prove that custom software is justified; the second is to keep ownership and operating cost clear.
Direct answer
Custom software is worth building when the workflow differentiates the business, existing products create costly workarounds, or owned logic and data access matter. It is usually a poor choice for a standard problem already solved well by maintained software. The decision should compare lifetime cost, control, integration risk, change frequency, and the consequence of depending on a vendor.
When this helps
What the work covers
The deliverable includes the decisions and operating path around the code, not just a repository.
Users, workflow, constraints, risks, and the smallest release worth funding.
Interface, data model, integrations, permissions, and technical trade-offs.
Full-stack application, tests, deployment, analytics, and release support.
Documentation, source ownership, operational access, and a clear maintenance plan.
Compare the options
| Option | Best when | Trade-off |
|---|---|---|
| Existing SaaS | The process is standard and configuration covers most needs. | Vendor limits and recurring cost. |
| Low-code | The workflow is internal, bounded, and changes frequently. | Complexity can outgrow the visual layer. |
| Custom software | Owned logic, integration, experience, or scale creates advantage. | Requires product ownership and maintenance. |
| Hybrid | A standard core can be extended through APIs and a custom interface. | Integration boundaries must be monitored. |
How the work runs
Observe users, data, constraints, and the cost of the current workaround.
Separate what should be bought, configured, integrated, or built.
Deliver one end-to-end workflow with tests and real users.
Document ownership, monitoring, releases, support, and the next product decision.
Relevant work
The work archive includes the 42 Network student intranet, 42 Lisboa launch systems, Agentic Brief, smart-contract launches, and Bitcoin inscription infrastructure.
FAQ
The contract should state ownership, licences, third-party services, credentials, and handover. I make those boundaries explicit before implementation.
Yes. A hybrid build around reliable APIs is often better than replacing everything.
After a short discovery, I estimate a defined release with assumptions, exclusions, dependencies, and risk. Large unknowns should become a paid discovery rather than false precision.
The proposal defines warranty, monitoring, support, maintenance, and handover. Continuing work is optional and should have a clear operating scope.
Yes, after a technical audit. The result may be stabilisation, staged replacement, or a smaller intervention rather than a rewrite.
Start with the constraint
I will help separate configuration, integration, and custom development before proposing a build.
Discuss the software