Evren Pala

The historical communication traffic search (HTS) system is used by legal authorities in Turkey. It collects call detail record (CDR) data from service providers and hosts query services to legal authorities. It comprises a database cluster, an ETL module, query services, and a data presentation module. The application runs in multi-thread and multi-instance mode to handle 10s of TBs of data flow per day. The project was taken over with some performance and inconsistency issues, causing frequent restarts. Problems have been resolved by redesigning the database cluster, re-implementing query services, and refactoring the ETL and presentation modules.

A web-based application that enables authorized government agencies to send location-based emergency messages to mobile subscribers. An emergency warning message can be SMS, short audio record, or cell broadcast message. The system is integrated with mobile network operators. Warning messages delivered through them. The application has a very limited number of users and uses multi-factor authentication.

A multithread, multi-instance application that is used for provisioning the services of mobile subscribers to the HLR device in the core network. System performance requirements were strict about responding to several millions of transactions each day. I was involved in the extension of the provisioning application for the 4G evolution process and implemented adapters for the new HSS device in the 4G network. I further supported the system with a 24/7 on-call service. The client (Turkcell), is one of the largest mobile network operators in Turkey.

This is a web-based customer relationship management application from Vodafone TR, one of the largest mobile network operators in Turkey. The application is used by all dealers and branches of Vodafone TR. I developed new features, fixed bugs, and provided 24/7 on-call support.

The Toptal React Academy is an exclusive learning program that teaches the React framework to select members of the Toptal network. After a month of study, all graduates are tasked with completing a 30-40 hour final project to build and deliver a React app from scratch. The above is a walkthrough of my final project.

A machine learning-based churn prediction and campaign recommendation application. As a back-end developer and solution architect, I:

• Designed and implemented a REST API.
• Implemented integration with a machine learning module.
• Implemented event stream consumers to collect raw events.
• Designed the database schema.
• Documented the project.
• Deployed the application on AWS from scratch. The tech stack includes React, Java, Python3, MySQL, Cassandra, and Spark.

A Java and Spring-based application in a microservice architecture. As a back-end developer, I implemented microservices communicating through SQS queues. Third-party API integrations and Kafka producers and consumers are implemented to fill out email content from templates.

A web application for scheduling appointments for pediatric behavioral healthcare services, where I worked as a back-end developer and implemented a GraphQL API using Kotlin, Spring Boot, and Hasura; also, I implemented scheduled jobs and 3rd-party API integrations and improved CI/CD pipelines.

A comprehensive Spring Boot-based data quality monitoring and observability platform built with Kotlin and Java 21.

The system provides real-time data quality assessment across multiple data sources, including PostgreSQL, Oracle databases, Kafka topics (JSON/Avro/Protobuf), and file systems (SFTP/FTP).

Features include automated schema discovery, configurable validation rules (regex, range, uniqueness, null checks), incident management with alerting, and multi-tenant architecture.

It implements OAuth2/JWT authentication via Keycloak, uses RabbitMQ for async processing, Redis for caching, and jOOQ for type-safe database operations, and supports large-scale data processing with virtual threads, distributed locks, and producer-consumer patterns for handling files up to 900 GB.

Built with Docker containerization and comprehensive CI/CD pipelines, it follows enterprise-grade patterns, including Flyway migrations, multi-module Gradle architecture, and extensive integration testing.