Islam Aleraky

• Origin: Implement a voice assistant system for education, taking ownership of the whole backend, frontend, and cloud architecture. Implementing 14 services with their infrastructure
  
• Implement operation readiness of each service: Before each service can go to production to service real users, It has to: have test coverage of more than 85%, good logging, metrics and alarms for each endpoint, global error
  handling, scale up and down based on a number of users/load.
• Implement load testing framework: Creating a distributed load test on the whole backend end to end to find the bottlenecks of the architecture and test different loads on each service and removing the bottlenecks and was able to scale 10X faster from 100 thousand to 1 Million requests per minute and reduce the cost of AWS from 20000$ to 2500$ per month.
• CMS/Contentful integration: writing Javascript/Typescript extensions for contentful(CMS system) using React to add more validation over the content

Flare

• Implement an Accounting System that handles billions of accounting transactions every day to be booked in the accounting ledgers and create an algorithm to speed up the bulk transactions processing by 10X times using in-memory trees.
• Implement Release Testing mechanism: built-in release testing mechanism with every commit to the accounting transactions to insure the coverage is above 90% and the tests are running successfully.
• Marketplace launch: Responsible for launching Amazon to 5 marketplaces (Belgium, South Africa, Nigeria, Chile, Colombia) and help in speeding up the release date by 10 weeks - managed to create a generic automation script that launched for the 5 marketplaces at once.

• Refactor to microservies: started the initiative to Refactor the monolithic app to microservices architecture to increase the development speed and limit the dependencies between development teams.
• IAM service: build an IAM service for authentication and authorization for microservices and enabling roles for users across services.
  • Public API: Creating a Public API project for Incorta to allow Customers outside of the system to query insights inside Incorta and make the endpoints handle requests that have data with over a billion records using streaming.
• Loading Phase Enhancing: Enhanced the loading data mechanism into Incorta by implementing a distributed Locking mechanism where each schema needs to write data on itself and read data from its dependencies by using the leader election mechanism
• Reduce regressions and help in refactoring: by increasing the code testing coverage in some packages from 6% to above 87%