Brussels Software Studio & Technology Publication
The Studio
The Weekly Dev designs and builds production software in TypeScript. The engagement is deliberately small: you talk to the engineer who writes the code, for the whole project.
The Weekly Dev SRL · Brussels, Belgium · Working across the EU
Most software work is sold by companies that do not write the software. A salesperson scopes the project, an account manager relays your questions, and the code is written by whoever is available. The gap between what you described and what gets built is where budgets go.
The Weekly Dev is structured to remove that gap. It is a Belgian SRL operated by one senior engineer. The person who scopes your project is the person who designs the data model, writes the code, and deploys it. When you ask why something works the way it does, you get the actual answer.
That structure has limits, and they are worth stating plainly: this studio is not the right choice for work needing a team of ten in parallel, twenty-four hour support coverage, or a large enterprise procurement process. It is a good choice when the work needs judgement, continuity, and someone who will still understand the system in two years.
Subscription applications with tenancy, billing, and an admin layer, built to take a first paying customer, not to demo.
Portals, booking systems, dashboards, and data-heavy interfaces where users log in and records change.
The systems that replace a critical spreadsheet or a manual process nobody has time to document.
Typed, documented interfaces between systems that currently exchange data by export and import.
Incremental replacement of systems that still work but can no longer be maintained or extended.
Architecture reviews, codebase audits, and roadmap direction where the deliverable is a decision, not code.
01
Take a product idea to a working, billable SaaS application, with the authentication, tenancy, and billing plumbing done properly the first time.
02
Applications, not brochure sites: software where users log in, data changes, and correctness matters more than the animation on the hero section.
03
Business software built around how your organisation actually works, instead of reshaping the organisation around off-the-shelf software.
04
Replace the shared spreadsheet, the manual copy-paste, and the process that only one person knows how to run.
05
Architecture reviews, codebase audits, and technical direction for teams that need a straight answer before committing budget.
06
Development capacity for design, branding, and marketing agencies, delivered behind your brand, without approaching your client.
One stack, chosen because it holds up over time, and because a single engineer can own all of it without the seams showing.
TypeScript
One language from the database schema through the API to the interface. Renaming a field becomes a compile error rather than a production bug.
React · Next.js · Tailwind CSS
Server-rendered by default, so pages load fast and are indexable. JavaScript is shipped to the browser only where the page is genuinely interactive.
PostgreSQL · Prisma
Relations, constraints, and transactions enforced by the database itself. Application code is not the last line of defence for data integrity.
tRPC · REST
Internal calls use end-to-end type inference; anything a third party consumes gets a documented, versioned REST interface.
Sessions · OAuth · Role-based access
Authentication, password recovery, and permissions built once as infrastructure, instead of re-checked ad hoc in each feature.
Docker · CI/CD · Cloud hosting
Every change moves through the same pipeline to a preview environment before production. Releases stop being events.
Scheduled jobs · Integrations · AI where it earns its place
Background work, system-to-system synchronisation, and, for genuinely unstructured input, a model in the loop. Not on principle.
A conversation about the problem, not the technology. What the software has to do, who uses it, what happens today, and what a good outcome looks like in six months.
Feasibility, risk, and the decisions that will be expensive to reverse. Sometimes the honest conclusion is that the project should be smaller, or that an existing product already solves it.
A written scope with an explicit list of what is excluded, alongside the data model, integration points, and infrastructure plan. Both sides agree what done means before work starts.
Work ships continuously to a live environment. You use the product while it is being built, and changes in direction happen while they are still cheap.
Deployment, monitoring, and handover documentation. Most engagements continue at a reduced cadence afterwards, but the codebase is written so that another team could take it over.
The first conversation costs nothing and is genuinely useful even if it ends in a recommendation to do something else.