Services
What we take on.
The work spans the whole stack. In practice a single engagement usually touches most of these.
Frontend engineering
React on the web and React Native on iOS and Android, sharing code across the three where it pays to rather than forcing it. Interface and interaction, state management, offline and poor-connection behaviour, and the accessibility and performance work that decides whether an application is actually usable rather than merely complete.
Typical work: building an application front to back rather than to a handed-over API; taking a design system into a working interface; fixing an app that works on the developer's machine and falls apart on a three-year-old Android phone.
Backend engineering
Designing and building services in Java and Spring Boot — REST APIs, relational data models on PostgreSQL, message-driven processing, and the integration work between systems that were never designed to talk to each other.
Typical work: a new service from data model to deployment; breaking a monolith into something maintainable; making an existing system correct and observable under load it was not built for.
Cloud architecture
AWS infrastructure defined as code — microservices and serverless components, API gateways, queues, managed databases and object storage. The emphasis is on choosing an architecture proportionate to the actual problem, because most systems are asked to carry far more complexity than their workload justifies, and it is expensive in ways that only show up later.
Typical work: designing infrastructure for a new service; migrating workloads to managed components; setting up deployment pipelines, backups, monitoring and the operational routine around them.
Fintech and payments
A decade of the work has been in financial services and payments, including cross-border money transfer. That domain brings requirements most systems never face: correctness that cannot be approximated, auditability after the fact, reconciliation, and failure modes where retrying is the wrong answer.
Typical work: digital wallets, payment and transfer flows, FX, ledger management, receivables and settlements; integrations with banks, payment partners and AML and sanctions-screening providers; automating reconciliation and transaction matching so financial controls stop depending on someone remembering to check; and reviewing a design where the cost of being subtly wrong is measured in money rather than downtime.
Product development
Complete applications rather than backend services alone: React Native clients on iOS and Android, a React web application, and the multi-tenant API and database serving all three. Including the parts that are easy to underestimate — authentication and role-based access, media upload and delivery, push notifications, transactional email, and the release process for getting builds through both app stores.
Background
AA Digital Services is led by Aris Apostolakis — over a decade of software engineering, currently working on cross-border payments at Hubpay, with earlier work for Nutmeg, Accenture and BMJ Publishing Group. MSc from the University of York, awarded with Distinction. Oracle Certified Professional, Java SE. Professional Scrum Master I.
Where a project needs more capacity than one engineer, we bring in collaborators we have worked with before rather than subcontracting to strangers — and you are told who is doing what.
Full history on LinkedIn.
Engagement terms
Ongoing retainer
A predictable monthly commitment. The best fit for product work that continues past a launch, which is most of it.
Fixed-scope project
A defined piece of work with a stated outcome. Suits well-understood problems where the boundary is genuinely clear.
Advisory
A smaller recurring commitment for review, architecture input and decision support, without day-to-day delivery.
Invoicing
Invoiced from Estonia in EUR. VAT handled per EU rules — intra-EU business customers with a valid VAT number are reverse-charged.
Is this a good fit?
Not every project is one we should take. These are the signals that an engagement usually goes well:
Likely a good fit
- The software has to be correct — money, personal data, or a regulated domain.
- You want one engineer accountable across the stack, not a layer to coordinate.
- The work continues past a launch rather than ending at handover.
- You would rather hear an honest "that is not worth building" than a quote.
Probably not
- You need a large team scaled up quickly.
- The scope is fixed, the deadline is fixed, and neither can move.
- You are looking for the cheapest possible build.
- The main need is design or marketing rather than engineering.