Yasser Aboudkhil

This is a simple gRPC service written in Go (Golang) which uses AWS KMS service to encrypt and decrypt whatever content you send. To make itself useful as a service for handling PII, it also acts as a dictionary of sorts for sensitive data. Essentially, consuming services will call it to encrypt a piece of data, and it will store sufficient unidentifiable metadata for performing decryption at a later stage when requested.

It uses AWS KMS but employs the envelope encryption technique recommended by Amazon to keep sensitive data encrypted and not rely on a central encryption key. This helps avoid storing the plain version of the encryption keys used (in case of a leak).

This service is strongly opinionated because it limits the radius of key exposure by enforcing a new encryption key per data entry encryption request. It is also opinionated due to the use of AWS KMS.

After creating this service, it has been key for my usage across many projects that I've worked on in the past, as it rendered the handling of PII (personally identifiable information) trivial and got the job done from a compliance perspective.

I designed, architected, and implemented a fintech operating system supporting robo-advisory, wealth management, digital banking, retail stock screening, trading services, and insurance and inheritance management.

The tech architecture is microservice-based. I built a set of microservices out of every specialized task (whether it is to perform sign-up and login authentication, KYC, funding, information retrieval or financial instrument information lookups, etc.). Wahed's backend is purely event-sourced, allowing all of its microservices to be asynchronous in nature. The services in the platform do not share any state and keep no session data, which makes it very easy and safe to scale the services independently and to recover from any system failures. Since Wahed’s microservice-based architecture doesn't rely on a single service but a set of them, it helps remove bottlenecks, as computationally consuming services are autoscaled.

The majority of the back-end services are created using Go, a proven, very mature language for creating rich and highly concurrent applications. It provides a robust, highly scalable, and very secure environment to create enterprise-grade server applications.