Iliyan Kostov

My first project is a software company, that provides global distribution service to airline companies. These distributions of data products are in the area of the travel industry.

The team I'm in is in the department of ticket reservation. The role I'm currently working is platform engineer and the objective is to maintain,
operate and develop the data processing pipeline, that is responsible
for ingesting the raw data, that's sent by the global distribution system,
process it through multiple layers and finally send it to the customer via various transfer protocols - FTP or S3.

• My initial project focused on the development of Advanced Driver Assistance Systems (ADAS) for a multinational automotive company. Within this project, my role was working as a DevOps Engineer on AWS cloud infrastructure management, where we were responsible for maintaining the entire cloud platform, including the deployment of customer products with the help of AWS ADF framework, AWS Account-Vending-Machine and AWS Control Tower custom REST API.
• We were also responsible governing the cloud platform by using AWS Cloudformation Policy templates and all of the essential resources. The automation of these processes was achieved through the use of AWS CDK Infrastructure as Code (IaaC).
• As part of my contributions to client services, I implemented the following solutions: - budget cost alerting on the entire platform with predefined % of exceeded budget to fire off email messages to the Business Owners - creating additional REST API endpoints for the Control Tower for specifying which newly created AWS accounts belong to which VPC. In there I also wrote the logic where a Lambda Function checks the type of account and based on which VPC template it belongs to - deploys the AWS CloudFormation StackSet to the appropriate account. - creating a High level design of data flow and replication for the entire Data platform
• My second project revolved around delivering valuable insights to enhance the shopping experience for customers and establishing loyalty programs for a renowned Whole Foods store company.I was positioned as Data Engineer on the Azure platform for ensuring data integrity and data validation on the Raw to Source to Prepared Layer. For data validation processes, we developed a Test Automation Framework built upon the pytest framework. The data flow initiates as follows:
• 1. A single entry of test data (.avro file) and a configuration file are copied from raw layer on Blob Storage
• 2. From that event of the configuration file being copied, it invokes an ADF (Azure Data Factory) pipeline, that in turn parses this test data towards Azure Databricks cluster
• 3. That cluster was responsible for doing minor transformations of the data, like converting the data type from avro -> parquet, changing column names for compatibility purposes, etc
• 4. After the transformations were done, the transformed test data gets copied to the Source layer of Blob Storage.
• 5. Test Framework via the Azure SDK is validating the data on the raw and source layer, for example: schema checks, data checks, duplication checks, column checks
• 6. After all tests are gathered, pytest creates a test execution results file junitxml_report.xml, thats required for the reporting tools to visualize the test results
• 7. We provided 2 different reporting tools: Zephyr Scale Rest API and Allure

• The intership consisted of mainaiting a legacy banking system for Balkan clients. The system was comprised of a desktop Windows XP compatible application, written in CAVO (Visual Objects) programming language.