Launching Online Lending in Romania in 3 Months
The goal was to launch a working online lending flow fast, under the client's brand, using the existing platform as the core.
Read More-
Services:
Integration, Custom Software Development, QA and Testing -
Industries:
Fintech, Lending -
Technology Stack:
REST API, Rebus, MS SQL, Azure Service Bus, Kubernetes, Jira, Confluence, Miro -
Project Summary:
A long-term client of Polytech Software operates a cloud-based digital lending platform that we have been designing, evolving, and supporting for more than 10 years. When the client decided to expand into Romania with a local partner, we launched a working online lending flow in about 3 months using the existing platform as the core.
Overview
A long-term client of Polytech Software operates a cloud-based digital lending platform that we have been designing, evolving, and supporting for more than 10 years. Over that time, we built not only the platform itself, but also the delivery engine around it: a team deeply familiar with the domain, stable release processes, integrations, manual QA, and automated testing.
When the client decided to expand into Romania with a local partner, the business opportunity was clear. The partner already had an established offline lending business and AI-powered consultant bots for customer communication. But there was a major growth limitation: new customers could not apply for loans online, and the product offering lacked short-term loans.
The goal was not to build a new lending product from scratch. The goal was to launch a working online lending flow fast, under the client's brand, using the existing platform as the core.
That is exactly what we did.
The Business Challenge
The Romanian partner had real market presence, existing customer channels, and operational processes. But its digital lending capabilities were limited.
The business needed to solve several problems at once:
- enable online lending for new customers
- extend the product line with short-term loans
- integrate digital decision-making into the partner's existing workflows
- avoid the time and cost of building a separate customer portal
- go live quickly enough to capture market momentum
This was a classic speed-to-market challenge with real operational and regulatory complexity behind it.
What We Were Asked to Deliver
We framed the engagement as an integration-led market launch.
Instead of rebuilding the partner's customer-facing experience, we connected our existing lending platform and risk decisioning technology to the partner's current infrastructure, processes, and communication channels.
The target solution included:
- our backend as the core lending engine
- our risk model as the core decision-making layer
- integration into the partner's workflow for pre-screening, KYC/AML, and application handling
- integration with Romanian government registries and local partner services for risk data enrichment
- integration with the partner's AI bots, so customers could go through the application flow in familiar channels without a new front-end build
This approach reduced time-to-market, limited delivery risk, and avoided unnecessary product duplication.
Our Approach
- 1
Fast discovery with business and technical alignment
We started with a focused discovery phase of about two weeks.
During this stage, we assessed:
- the partner's business processes
- the existing IT landscape
- integration points and operational constraints
- local legal and compliance considerations around data sources, customer consent, and data storage
This gave us a realistic implementation map early, before delivery risk had a chance to accumulate.
- 2
Integration architecture before feature expansion
We defined the target integration architecture, data mapping for the risk model, ownership boundaries between teams, and a phased rollout plan.
The delivery strategy was clear:
- launch an MVP online lending flow for new customers first
- then expand with additional data sources, risk rules, product variations, and promotional mechanics
- 3
Delivery with the existing Polytech team
We did not assemble a new team for this launch.
The project was delivered by our existing team already working on the platform and fully familiar with the architecture, business logic, release process, and quality controls. That removed onboarding delays, reduced knowledge-transfer risk, and allowed us to move straight into execution.
This is one of the biggest advantages of working with an experienced product team around an established platform: speed comes from familiarity, not from rushing.
What Polytech Software Delivered
End-to-end systems integration
We connected the partner's offline processes and digital channels with our lending platform to support the full application lifecycle:
application → verification → decision → disbursement → servicing
The result was a functioning online lending flow embedded into the partner's operating model, not isolated from it.
Risk data integrations for Romania
We integrated Romanian government and partner services, where legally permitted, to enrich decision-making and support:
- KYC/AML checks
- identity verification
- credit history checks
- income and profile validation
This allowed the lending flow to operate with stronger data inputs and more reliable automated decisioning.
Risk model execution in our backend
We placed the risk rules and decisioning model inside our backend services, with clear traceability of inputs and outputs, auditability, and the ability to tune decision logic without rebuilding the front-end experience.
That mattered for both business agility and operational control.
AI-assisted UX instead of a new portal
Rather than building a separate online portal, we integrated our APIs into the partner's AI bots and existing web channels.
This gave end users a simpler and more familiar experience while dramatically reducing launch overhead. It also helped the client reach the market faster and spend less on front-end development that was not essential to validating demand.
Process structure and delivery governance
We translated business requirements into a delivery-ready format, documented scenarios and non-functional requirements, and aligned responsibilities across teams.
This included:
- SLA expectations
- availability requirements
- logging and observability expectations
- ownership split between both sides
- integration responsibility mapping
Quality engineering and release control
To keep speed from turning into instability, we reinforced the delivery with strong QA and release discipline.
We implemented and maintained:
- end-to-end tests for critical flows
- automated regression coverage
- integration seam checklists
- control of data consistency across services
- QA gates around releases
This helped protect release velocity without sacrificing stability in a high-impact business flow.
Cross-team execution
We ran regular working sessions with business stakeholders, analysts, and development teams from both sides. That allowed us to align requirements quickly, validate integration schemes, and resolve changes without long delays.
For projects like this, delivery speed is not just about engineering capacity. It is also about decision-making flow. We kept that flow moving.
Timeline and Outcome
Discovery
Focused assessment of business processes, IT landscape, and compliance
Implementation to launch
From implementation plan to go-live
Total timeline
Full path from assessment to go-live
After the Launch
The solution launched on time in the Romanian market under the client's brand.
After the launch:
- the AI assistant used by end customers was able to call our backend services and risk model in real time
- online lending became available to new customers for the first time
- the product line expanded with short-term loan offerings
- the partner activated promotions post-launch
- customer volume grew and market position strengthened
Most importantly, the client achieved digital market entry without building an entirely new product stack.
Why This Worked
- 1
A proven platform and a team that already knew it
Because we had over 10 years of experience building and supporting the lending platform, we could start from a position of operational maturity. No ramp-up. No re-learning. No avoidable architecture mistakes. - 2
Integration instead of unnecessary rebuilding
We did not treat the challenge as a greenfield product build. We used what already worked: the platform, the decisioning engine, the partner's channels, and the AI-based customer interface.
That decision saved time, cost, and delivery effort.
- 3
Controllable, auditable risk decisioning
Placing the risk model in the backend gave the client a more manageable and transparent decisioning layer. The business could evolve rules faster, with traceability and less dependency on front-end changes. - 4
Strong quality discipline under delivery pressure
Fast launch projects often break when teams cut corners on testing and release control. We did the opposite. Automated regression, end-to-end coverage, and QA gates helped us move quickly without losing control over stability.
Tech Stack and Delivery Practices
Integrations and messaging
REST API, Rebus
Data and infrastructure
MS SQL, Azure Service Bus, Kubernetes
Process and collaboration
Jira, Confluence, Miro
Delivery practices
Agile / Scrum, end-to-end testing, automated tests, QA gates, traceability for risk model decisions
Business Impact
This project is a strong example of how Polytech Software helps lending and fintech businesses launch faster by combining:
- an existing production-grade platform
- deep domain expertise
- integration-first thinking
- controlled risk decisioning
- quality-focused delivery
Instead of spending months building a new customer portal and re-creating core lending logic, the client used a platform and team already capable of supporting the business model. That shortened time-to-market, reduced implementation risk, and created a commercially usable result in about three months.
Conclusion
For companies in lending and regulated digital finance, speed matters, but speed without control is dangerous.
This launch succeeded because we did not approach it as a rushed custom build. We approached it as a structured market-entry integration using a platform, team, and delivery model already proven in production.
The result: a new online lending capability launched in Romania in about 3 months, under the client's brand, with lower cost, lower risk, and faster commercial impact.