Share

Client

The client is a retail and consumer analytics provider working with large-scale data to support personalized digital media campaigns across online and in-store channels. Its platform processes data used to deliver targeted campaigns and support different advertising scenarios.

The company operates a big data environment that handles large data feeds and millions of daily impressions. As the need for new campaign types and greater processing capacity increased, the existing platform required modernization to improve performance, scalability, and reliability.

Partnership goal:
  • Modernize the big data platform. Migrate legacy orchestration and processing components to modern data engineering tools while improving platform performance and reliability.
  • Support scalable personalized campaigns. Improve data processing and accessibility to accelerate the delivery of new campaign types across online and in-store channels.

Challenge

The existing big data platform relied on legacy components that limited flexibility and made it more difficult to scale personalized campaigns across digital and in-store channels.

The client needed to improve data processing performance, reduce time-to-market for new campaign types, and modernize orchestration and engineering workflows without disrupting existing data operations.

The project also required better data access and search capabilities, along with a more reliable infrastructure capable of supporting large-scale data processing and future growth.

Solution

What We Built

Techbar developed and maintained the core big data platform supporting personalized media campaigns, modernizing its data processing and orchestration architecture.

A key part of the project involved migrating workflow orchestration from Oozie to Apache Airflow to improve flexibility and reliability. Azure Data Factory and Airflow supported pipeline orchestration, while Apache Spark, Databricks, Hive, and NiFi were used to optimize data processing.

The platform incorporated Azure Data Lake Storage (ADLS) and HDFS for data storage, with Hadoop and distcp supporting distributed data processing and transfers. Apache Solr provided search and indexing capabilities to improve data access and campaign performance.

CI/CD pipelines were established through Azure DevOps, with automated testing and monitoring integrated into the engineering workflow.

How We Worked

The project followed an Agile delivery process involving cross-functional engineering teams responsible for platform development, migration, and performance optimization.

The team focused on reducing operational risks during the transition from legacy components, with particular attention to migration reliability and processing performance.

End-to-end DevOps practices, automated testing, and monitoring supported ongoing development and maintenance of the platform.

Results

The modernized platform enabled the client to introduce new personalized campaign scenarios across online and in-store channels, contributing to measurable revenue growth.

Optimized processing and improved data access supported large-scale data feeds and millions of daily impressions, while the updated architecture improved scalability and system resilience.

Migration to modern orchestration and data engineering tools also established a more flexible foundation for future campaign development and platform growth.

More Client StoriesView all case studies

Do Not See Your Scenario Here?

Most of our work stays under NDA. Describe your engineering challenge and we will send relevant examples along with pre-vetted candidate profiles within 48 hours.

Dedicated software engineers