000 — BACKEND SYSTEMS

Backend systems that stay up when the business scales.

We design, build and fix backends: APIs, integrations, data pipelines and cloud infrastructure, plus refactoring of legacy code and technical debt. For product companies and operations-heavy businesses whose software has outgrown its first version. Work runs in fixed-price 2-week sprints from €4,900, starting with a €690 technical audit when the system already exists.

Start a project

001 — Sound familiar?

The backend is where the real problems live.

  • Integrations with your ERP, CRM or payment provider fail silently, and someone fixes the data by hand on Monday.
  • One slow query takes the whole app down at peak hours, and nobody knows which one.
  • Every feature takes twice as long as estimated because the codebase fights back.
  • The cloud bill grows faster than your users, and nobody can explain the line items.
  • Auth, permissions and secrets were "temporary" two years ago. A security questionnaire from a big customer is now sitting unanswered.
  • When something breaks, the first person to notice is a customer.

002 — What's included

What we take on

Tech debt amounts to 20–40% of the value of the whole technology estate, and 10–20% of new-product budgets get diverted to it (McKinsey). We build new systems so they do not start that way, and pay down debt in existing ones while features keep shipping.

  • APIs and services

    REST, GraphQL or typed RPC APIs with versioning, validation, rate limits and documentation generated from the code. Background jobs and queues for anything that should not block a request.

  • Integrations

    ERP, CRM, accounting, payments, shipping, email. Idempotent webhooks, retries with backoff, dead-letter queues and an audit trail, so a failed sync is visible and recoverable instead of silent.

  • Data and databases

    Schema design, migrations under version control, indexes that match real queries, backups with tested restores. PostgreSQL by default; Convex or Supabase when real-time and speed of iteration matter.

  • Legacy refactoring and tech debt

    We replace legacy systems incrementally, strangler-fig style (Fowler): new code grows around the old, one route or module at a time, with production traffic proving each step. Never a big-bang rewrite.

  • Security

    Auth and sessions, role-based access, Row Level Security, secrets in a vault, dependency scanning and audit logs. We build SOC 2 / ISO 27001-ready: access control, encryption, backups and documented incident response.

  • Performance and cloud cost

    Profiling, query plans, caching and right-sized infrastructure. We find what is slow and what is expensive, show you the numbers, and fix the parts that matter first.

  • Infrastructure and observability

    Infrastructure as code, containerised services, CI/CD with rollback, and structured logs, traces and metrics wired to alerts. You learn about problems before your customers do.

Who we build backends for

  • CTOs of software companies who need senior backend capacity without hiring a team.
  • Operations-heavy businesses running on integrations between ERP, e-shop, warehouse and accounting.
  • Startups whose MVP backend — AI-built or not — is buckling under real users. See vibecode rescue.
  • Teams with a legacy system that must keep running while it is being replaced.
  • Companies facing a security review, due diligence or an enterprise customer's questionnaire.

003 — How it works

How a backend engagement runs

  1. Technical audit

    For existing systems: architecture, data model, dependencies, security, performance hot spots and cost drivers. A written report with a prioritised plan. Fixed €690, credited to the first sprint.

    Day 1–3

  2. Architecture and plan

    For new systems or big changes: a short design doc, ADRs for the key decisions, and a sprint plan with fixed prices. You approve before we write production code.

    Week 1

  3. Build in sprints

    2-week sprints with a fixed scope. Every change goes through CI, a preview environment and a review. Weekly demo, written progress note.

    Week 2+ (per 2-week sprint)

  4. Production rollout

    Feature flags, gradual rollout, migrations that can be reversed, and dashboards watching the change. Old and new run side by side until the numbers say it is safe.

    Per release

  5. Operate

    Runbooks, alerts and on-call handed to your team, or kept by us on a care plan with an SLA.

    Ongoing

004 — Pricing

Backend pricing

Audit first if the system exists, then fixed sprints. If you need us long term, a care plan keeps the system healthy between projects.

  • Technical audit

    €690

    fixed · 3 business days

    • Architecture, data and dependency review
    • Security and access-control findings
    • Performance and cloud-cost hot spots
    • Prioritised plan with sprint estimates
    • Fee credited to the first sprint

  • Build sprint

    Most chosen

    from €4,900

    per 2-week sprint

    • Fixed scope agreed up front
    • APIs, integrations, data or refactoring work
    • Tests, CI gates and preview environments
    • Observability on everything we ship
    • 30-day warranty on shipped code

  • Care Plus

    from €1,490

    per month

    • 99.9% monthly availability target with service credits
    • Priority incident response
    • Continuous tech-debt paydown
    • Monthly DORA metrics report

Every quote is fixed before work starts. If we overrun our own estimate, that is on us, not you.

005 — Definition of done

Definition of done for backend work

Stack

  • Node.js
  • TypeScript
  • PostgreSQL
  • Convex
  • Supabase
  • Redis
  • Docker
  • Terraform
  • OpenTelemetry
  • Sentry

definition_of_done.md

  • Tests on critical paths

    Unit tests on business rules, integration tests against a real database, contract tests on external APIs.

  • CI gates and preview deploys

    Type checks, linting, tests and a build run on every pull request. Nothing merges red. Every branch gets a preview deploy you can click before it reaches production.

  • Observability built in

    Structured logs, traces and metrics on every service, with alerts on the symptoms users feel: errors, latency, failed jobs.

  • Security review

    Every change touching auth, permissions, data access or secrets gets a dedicated security review before merge.

  • Reversible by default

    Migrations with a down path, feature flags on risky changes, and a rollback we have actually tested.

  • Docs, ADRs and runbooks

    Every non-obvious decision gets a short architecture decision record. Every operational task — deploy, rollback, restore, rotate a secret — gets a runbook someone else can follow at 3 a.m.

  • Your repo and cloud from day one

    Code lives in your GitHub organisation and runs in your cloud accounts from the first commit. Full IP transfers to you on payment. No hostage situations, ever.

007 — Questions buyers ask

Questions buyers ask

How much does backend development cost?

Work runs in 2-week sprints from €4,900, each with a fixed scope and price. For an existing system we start with a €690 technical audit, credited to the first sprint, so the estimate is based on your code rather than a guess.

Do you rewrite legacy systems?

Rarely all at once. We replace them incrementally — new code around the old, one module at a time, with production traffic proving each step. A big-bang rewrite stops feature work for months and tends to rediscover every bug the old system already fixed.

Can you work with our in-house team?

Yes, and that is often the best setup. We join your repo, your standups if useful, and your review process. We can own a well-defined area — integrations, a migration, observability — or pair with your engineers so the knowledge stays in-house.

Who owns the code, infrastructure and IP?

You do. Everything lives in your GitHub organisation and your cloud accounts from the first commit, and full IP transfers on payment. We use your accounts with access you can revoke at any time.

What if you disappear?

Then you still have everything: code in your repo, infrastructure as code, ADRs explaining the decisions and runbooks for every operational task. We build so that another competent engineer can take over without calling us. You can leave anytime, with a handover pack.

Do you sign an NDA and a DPA?

Yes. A mutual NDA before we see anything, and a GDPR Article 28 data processing agreement whenever we process personal data. We can also work under your existing security policies and access controls.

What timezone do you work in?

CET, Monday to Friday, in English, Slovak or Czech. That overlaps fully with European teams and covers the morning for US East Coast teams. Incident response outside business hours is available on Care Plus or a custom plan.

Can you reduce our cloud bill?

Usually, yes. The audit includes a cost review: oversized instances, chatty services, missing caches, logs nobody reads. We show you the numbers and the trade-offs, then fix what pays back fastest.

008 — Related services

100 — BACKEND SYSTEMS

Have a backend that needs to grow up?

Tell us what the system does and where it hurts. We will reply within one business day with a practical next step — usually a short call or a technical audit.

Start a project