Note
System assessment before commitment
Why a short system assessment helps before a team commits to a rewrite.
Open note ↗Work / Backend & platform engineering
Backend work across loan servicing and origination, where my scope grew from product delivery into system assessment, API gateways, cloud deployment, and integrations.
I joined InfinID on the loan-servicing platform after several years working in a more traditional banking environment. The stack, pace, and engineering expectations were different: services were smaller, integrations were more numerous, and cloud and infrastructure concerns sat much closer to everyday backend work.
My early work mixed feature delivery with questions that had to be answered before implementation. I assessed an open-source core-banking system and wrote an RFC covering its APIs, security considerations, latency, delivery requirements, and architectural and operational risks. I later worked on another RFC around consolidating two overlapping legacy user-management systems.
Those projects gradually pulled my work beyond implementing defined changes. I was increasingly being asked to understand how systems fit together, where their boundaries should sit, and what introducing or reshaping one component would mean for the rest of the platform.
I joined InfinID on the loan-servicing platform, in an environment that was faster-moving and more cloud and integration-heavy than the systems I had worked on previously.
My early work mixed feature delivery with questions that had to be answered before implementation. I assessed an open-source core-banking system and wrote an RFC covering its APIs, security considerations, latency, delivery requirements, and architectural and operational risks. I later worked on another RFC around consolidating two overlapping legacy user-management systems.
Those projects gradually pulled my work beyond implementing defined changes. I was increasingly being asked to understand how systems fit together, where their boundaries should sit, and what introducing or reshaping one component would mean for the surrounding platform.
A growing part of the role sat between backend engineering and platform concerns.
I built environment-specific tooling around KrakenD so gateway configuration and deployment could be handled consistently across environments, with logging, security, and observability included in the delivery model. The work connected backend implementation with configuration, deployment, and production concerns rather than treating the gateway as a standalone component.
Around the same period, I built a configurable loan-assessment playground for credit analysts. It let them experiment with parameters such as debt-to-burden and loan-to-value ratios without turning every scenario into a separate backend change.
The two projects were different, but both expanded the kind of responsibility I was taking on: not just implementing application logic, but building mechanisms that other people could use to work with the system.
Toward the end of 2024, I moved into a broader loan-origination and servicing revamp.
The work became more infrastructure-aware. I worked on Go microservices and API-gateway integration deployed on EKS, alongside loan-origination and document-management flows for agents and B2B partners.
I also refactored OCR-related services and integrated AWS Textract, worked on external integrations including Adjust, Freshworks, and espoCRM, and mentored junior engineers around SOLID principles and cleaner code structure. Key-Contributions
By this point, backend work meant thinking across a wider surface: service boundaries, external systems, deployment, configuration, observability, and how changes would behave once they reached a real environment.
InfinID was an important growth period, but not an uncomplicated one.
The later phase exposed gaps that my previous experience had not exercised as heavily: distributed-system context, testing depth, modern engineering practice, event-driven architecture, architectural intuition, and the speed required to build context across several moving systems.
The experience changed how I think about engineering experience itself. Years spent working and the composition of those years are not the same thing. I left with a clearer picture of the repetitions I was missing, the environments in which I still needed to grow, and the engineering standards I wanted to work toward. That recalibration became useful in the roles that followed.
02 / Go deeper
Artifacts owned by this role, with the problems, systems, and decisions they explore in more detail.
Note
Why a short system assessment helps before a team commits to a rewrite.
Open note ↗Case study
A closer look at the KrakenD work: how gateway configuration and deployment were structured across environments, and the trade-offs involved in making those changes repeatable and reviewable.
Open case study ↗Case study
An internal playground where analysts could test assessment-rule changes without production access or one-off engineer scripts.
Open case study ↗Project
CLI and templates for API-gateway configuration across environments, with container build and publish automation.
Open project ↗Sketch
A visual model of how structured KrakenD configuration moves from source templates through environment-specific builds and deployment.
Open sketch ↗Knowledge
A guide to naming which parts of a system can fail together and which consequences cross the boundary.
Open knowledge ↗Atlas context
See how this role connects to the projects, case studies, notes, and ideas around it.