Who owns it?
Maintenance, security response and commercial control matter after launch.
Technologies
Fit the system.
Fit the owner.
A stack is a long-term operating decision. We choose for the users, constraints and team that will live with it.
Choose for the operating life, not the demo
The stack is selected according to existing systems, team ownership, hosting constraints, hiring market, failure cost and the realistic exit from each provider. The list below describes technologies Firequark can work with; it is not a promise that every project needs them.
Working stack
Tools chosen
for the boundary.
| Layer | Common choices | Decision factors |
|---|---|---|
| Web | TypeScript, React, Next.js, standards-based HTML/CSS/JS | Interaction depth, accessibility, delivery profile |
| Services | Node.js, PHP, Python | Host, latency, team ownership, ecosystem |
| Data | PostgreSQL, MySQL, SQLite, Redis | Consistency, tenancy, scale, recovery |
| Mobile | Swift, Kotlin, React Native, installable PWA | Device capability, distribution, offline behavior |
| Infrastructure | Docker, CI pipelines, Linux, shared hosting, managed cloud | Cost, control, observability, operator skill |
| Integration | REST, webhooks, queues, file exchange | Provider limits, retries, idempotency, reconciliation |
Questions first
Before a dependency
enters the system.
Who can operate it?
The system must match the actual team and support path, not an imagined future organization.
How does it fail?
Timeouts, retries, partial results and unavailable providers are designed as states.
How do we leave?
Data export and replacement boundaries are understood before adoption.
Ignition
Keep the stack you can own.
We work with existing environments when they can safely support the required behavior.