Programmatic Video Distribution Platform

From Unscalable Legacy to Maintainable System: Programmatic Video Distribution Platform Revived

89% Reduction in Infrastructure Costs

Achieved by eliminating unnecessary resources and overprovisioning

Zero Downtime, Zero Data Loss

Flawless migration from legacy system to a scalable architecture

Future-Ready Foundation Established

The modernized system now supports ongoing product development and scalable growth

Industry

AdTech

Media

Programmatic Platform

Provided Services

Legacy Software Repair Service Digital Transformation Software Application Support & Maintenance

Duration

8 months + continuous support

Tech Stack

Ruby, Rails, Scala, Clojure, AWS, Terraform, ECS (Amazon Elastic Container Service)

[ about/ ]

Our client is a programmatic video advertising platform connecting brands, agencies, and publishers in one ecosystem.

Our client is a programmatic video distribution platform, with a Demand Side Platform (DSP) and Data Management Platform (DMP) for advertisers, as well as a Supply Side Platform (SSP) for publishers.

These tools allow advertisers to execute programmatic media buying and audience targeting, while publishers efficiently manage their video advertising inventory to maximize revenue.

[ challenges/ ]

The Core Challenges Solved

It was once an innovative and successful ad project, but became a legacy burden after multiple ownership and development team changes.

C1:

Broken Production Environment

The production environment was only partially functional—some core services weren’t running at all, making the system unreliable.

C2:

Outdated Tech Stack & Monolith Architecture

The platform ran on a 2011–2016 legacy stack with deprecated dependencies. The monolith made it hard to scale or refactor.

C3:

Undocumented Code & Unclear Configuration

No documentation on expected system behavior or configurations, making debugging and onboarding extremely difficult.

C4:

Infrastructure Waste

Infrastructure inefficiencies—high cloud spend caused by overprovisioning.

C5:

Unmaintainable Codebase

No ability to maintain or contribute to the code—even restarting the service was risky, with no guarantee it would come back up.

C6:

No CI/CD pipeline

Deployments were done manually, increasing the risk of errors and slowing down delivery cycles.

Talk to Our Engineers About Modernizing Your System

It’s kind of our thing

Book a Free Consultation

Tap the button

[ how we did it/ ]

Solution & Implementation

To stabilize, optimize, and modernize the platform, our approach focused on key infrastructure, deployment, and development improvements.

S1:

Infrastructure & Cost Optimization

Conducted a full infrastructure audit, eliminating unnecessary resources.
Optimized infrastructure, reducing operational costs by 89%.

Developed Infrastructure as Code (IaC) using Terraform, ensuring repeatability and scalability—critical for deploying new instances for partners and continuously improving the system.

S2:

Tech Modernization & Stability Enhancements

Built a CI/CD pipeline from scratch, enabling seamless deployments.
Containerized services, including the monolithic system and a database maintenance service.
Built a CI/CD pipeline from scratch, enabling seamless deployments.

Enabled environment-based configurations, adding flexibility for different deployment scenarios.
Modernized logging, improving debugging and observability.

S3:

Development Environment & Execution Roadmap