Software services
Integration & event architecture
The engineering behind most 'the platform is down' calls: integration architecture built around idempotency, bounded retries and dead-letter handling, event streams that scale, and reporting data that reconciles.
Outcomes
- Fewer incidents caused by systems you do not own
- Faster onboarding of new partners and data sources
- Finance and operations teams who trust the numbers
The challenge
What usually brings clients to us
Your platform depends on systems you do not control: a partner's API, a payment provider, an ERP, a customer's data feed. When one of them fails at 2 a.m., it is your name on the incident.
Most of these failures are ordinary: a retry loop with no backoff, a webhook with no idempotency key, a batch job nobody load-tested. They only appear once real traffic and real partners are involved, which is exactly when they are most expensive.
What you receive
- Integration architecture with idempotency, retry and dead-letter behaviour designed in
- Event-driven processing on Azure, sized for peak load
- Reconciliation and reporting pipelines
- Monitoring and alerting that surface partner failures before customers do
- Documentation of every integration contract and failure mode
Approach
How we deliver it
Design for the partner's bad day
Queues and event streams on Azure Service Bus, Event Hubs and Functions, built so a third party's outage is absorbed rather than passed on to your customers.
Contracts written down
Every integration gets a documented contract: payloads, limits, error behaviour and ownership. New partners become configuration rather than engineering.
Data that reconciles
Reporting pipelines and reconciliation jobs that can explain any record after the fact, including the unglamorous work of making an existing database stop being the bottleneck.
Observability as a deliverable
Tracing, alerting and dashboards that answer 'what happened?' in minutes, handed over as part of the work.
Client work
Where we have delivered this
Anonymised engagements delivered by the Codentures team. See all client work.
- Operations
Real-time operational intelligence API and data portal
A high-volume operations business
A real-time operational intelligence API and data portal, with deep query and index optimisation on an existing SQL Server estate.
- 4x query performance gain
- Live operational data portal
- E-commerceEU
Multi-vendor marketplace with automated payouts
A European retail group
A multi-vendor marketplace on Stripe Connect with automated seller payouts, payout reconciliation and real-time inventory sync.
- 40+ active vendors
- Automated payout reconciliation
Fit
Who this suits
SaaS and platform businesses whose revenue depends on third-party systems behaving well, and enterprises connecting a growing number of internal and partner systems.
For a single, low-stakes integration, a small fixed-scope piece of custom development is usually the better fit.
Typical technologies
- Azure Service Bus
- Event Hubs
- Azure Functions
- Logic Apps
- Azure Data Factory
- REST and webhooks
- SQL Server
- PostgreSQL
- .NET / C#
Send us the system and the symptom
A scoping conversation takes an hour. You will leave it knowing how big the problem is, where the risk sits, and what the sensible first step costs.