Infrastructure

Where we host the systems

We don't sell hosting. We pick it for the specific project — so your client's app runs fast, secure and without internal lock-in to a single provider.

For some projects Vercel + Railway is enough. For larger ones — dedicated OVH servers with our administration. Every choice is justified to the client's team and documented in an architecture decision record.

Where we host the systems
Stack

Our 4 hosting paths

Depending on scale, compliance requirements and client preferences — we combine these 4 approaches into hybrid configurations.

  • Vercel

    Edge & frontend

    Deployment of Next.js and Nuxt frontends on a global CDN. Edge functions, automatic SSL, preview environments per pull request, instant rollback.

    • Marketing sites and landing pages — instant scaling for traffic spikes
    • SaaS frontends with geo-distributed users (latency < 50ms worldwide)
    • Preview deploys per PR for fast iteration with the client
    • Edge functions for A/B tests and geolocation
  • Railway

    Backend services

    Hosting backends (Laravel, FastAPI, Node) with automatic deploy from git push. PostgreSQL, Redis and private networking out-of-the-box.

    • Laravel API + PostgreSQL backend without DevOps overhead
    • Worker queues + scheduled jobs (cron) — zero configuration
    • Staging environments in 1 click with database snapshot
    • Microservice architecture with private networking between services
  • OVH

    Dedicated VPS / cloud

    European hosting (Poland, France, Germany) — full machine control, GDPR-compliant, optimal price/performance ratio for databases and disk-heavy workloads.

    • Databases requiring Poland/EU data residency
    • Legacy apps requiring full SSH + custom kernel
    • Hosting reverse proxy (NGINX Proxy Manager) under our administration
    • Self-hosted GitLab / Sentry / Grafana for enterprise projects
  • Client's dedicated servers

    On-prem & private cloud

    The client has their own infrastructure (data center, private VPS, office server) — we deploy the app at your place, document procedures, train the team.

    • Apps with PII / medical data — zero exposure outside the client's network
    • NIS2 / KRI compliance / regulated industries
    • Integration with local systems (ERP, CRM) without VPN to public cloud
    • Client already has a server — wants to avoid double hosting costs
How we choose

Philosophy: stack to fit the project, not the other way around

  • 01

    Zero vendor lock-in

    Every stack is documented in an ADR — switching providers takes weeks, not months. Data, code, configuration — all yours, in git.

  • 02

    3-2-1 Backup

    Three copies of data, two different media, one off-site. RPO max 1h for prod, RTO max 4h. Restore test every quarter.

  • 03

    Monitoring from day one

    Sentry + structured logs + uptime check + SMS/email alerts. Client sees the same as we do — full system status transparency.

Have questions about hosting your project?

We'll help you pick the optimal stack — free audit of current infrastructure and a proposal for concrete changes.