01
We choose boring tools deliberately.
Postgres before Mongo. Hetzner before Lambda for steady workloads. Boring is fast to ship and slow to break.
04 / Practice
Core practice · Enterprise Systems
The boring engineering that runs your business.
We architect and build the complex software backbones that scale with serious teams: bespoke platforms, internal tools, integrations and replatforming of legacy systems.
01
What we build
Five capabilities. All shipped, all maintained.
Documentation-style. We list what we'd defend in a technical review — nothing else.
- 01
Custom platforms and back-office systems
Multi-tenant SaaS, marketplaces, internal admin and ops tools. TypeScript, Python and PHP — pragmatic, not religious.
- 02
API and integration layers
REST, GraphQL, gRPC. Webhook orchestration. Idempotency, retries and audit logs by default — not as an afterthought.
- 03
Data infrastructure and CDP
Postgres with read replicas, Clickhouse for analytics, Kafka or Redpanda for streaming. dbt for modeling and lineage.
- 04
Legacy replatforming
Strangler-fig pattern. Move off the monolith without a rewrite-from-scratch death march — and without freezing the product roadmap.
- 05
Multi-tenant architecture
Schema-per-tenant or row-level security per workload. Tenant-aware caching, per-tenant rate limits, per-tenant cost panels.
02
When to call us
Common scenarios.
If two of these sound like the room you're sitting in, we're probably the right call.
- 01You have a working product but the architecture is the bottleneck.
- 02You're replatforming and want to do it once, properly.
- 03You've hired developers but lack a senior architect.
- 04You need a complex integration delivered on a hard deadline.
- 05You're spinning up a new product line and want a senior team to anchor it.
03
How we work
Discovery, build, operate, hand off.
One team carries the engagement from first scope to production support.
01
Discovery — 3 to 6 weeks, flat fee
Architecture, scope, risks, costs. Output is a written plan with comparative trade-offs and an honest cost band.
02
Build — 12 to 40 weeks, fixed scope or T&M
Senior engineers, weekly demos from real branches, merges into main from week one. No three-month black box.
03
Operate — ongoing, monthly retainer
Observability, runbooks, on-call rota, capacity planning. Operations is part of delivery, not an upsell.
04
Hand off — when your team is ready
Documentation, ADRs, knowledge-transfer sessions. We leave a system your engineers can run alone.
04
Stack · our defaults
Boring tools, deliberately chosen.
What we reach for first. Stack shifts per engagement — we don't insist on a religion.
Backend
- TypeScript (Node, Bun)Default for new platforms
- Python (FastAPI)Data, ML and integration paths
- PHP 8 + SymfonyLegacy modernization
- GoPerformance-critical paths
Data
- PostgresOLTP default, read replicas, RLS
- ClickhouseAnalytics and event tables
- RedisCache, queues, rate limits
- Kafka / RedpandaStreaming and event sourcing
- dbtModeling, lineage, tests
Infrastructure
- AWS / GCP / HetznerPer workload economics
- Terraform / PulumiIaC, every environment
- Docker / KubernetesWhen the topology earns it
- GitHub ActionsCI, deploy, eval gates
- Sentry / Grafana / LokiErrors, metrics, logs
05
Architecture principles
The rules we apply, every engagement.
Four principles that guide how we make trade-offs. We've watched what happens when we don't.
02
Every system has a working version in main from week one.
We don't disappear for three months and present a finished build. You watch it grow, week by week, on a real branch.
03
Operations is part of delivery.
Observability, runbooks, on-call rota — included, not extras. The deliverable is software in production, not a hand-off.
04
We document the why, not just the what.
ADRs (Architecture Decision Records) for every non-trivial choice. Future engineers — yours or ours — read them and understand the trade-offs.
06
Selected work
Two engagements, briefly.
Both shipped, both maintained, both under reference if you'd like one.
07
Pricing & engagement
Three engagement models. Honest ranges.
Real scope and real estimate on a call. The bands below are where most engagements land.
Flat fee
Discovery
3–6 weeks
€20k–€60k
- Architecture and ADRs
- Risk register
- Replatforming options
Fixed scope or T&M
Build
12–40 weeks
€120k–€800k
- Senior engineering team
- Weekly demos in main
- Operations included
Monthly retainer
Operate
Ongoing
from €10k / mo
- On-call coverage
- Capacity planning
- Quarterly architecture review
08
FAQ
Five questions we hear most.
We have an in-house team. Can you augment?
Yes — pair-and-train model. We embed alongside your team, ship together, and write the kind of documentation your engineers can own after we leave.
Can you take over a legacy codebase?
Yes. We start with a 2-week audit: architecture map, risk register, replatforming options. You get a written plan before any code changes hands.
Will you sign IP assignment?
Yes — by default. Your code is yours. Our standard contract assigns all work product on payment.
Do you stay for hand-off after delivery?
Yes — included in every engagement. Runbooks, on-call coverage during the handover window, knowledge-transfer sessions with your engineers.
Compliance: SOC2, ISO 27001, HIPAA, GDPR?
Yes — supported on request, with documented controls on file or trackable in a shared compliance workspace. We've shipped under each of these regimes.
Have a platform that needs to scale, and an architecture that won't?
Send a short brief — current stack, the bottleneck, the deadline. We'll reply with concrete next steps.