Raisin retired Jenkins and built an autonomous, governed delivery engine that scales with the business.
Executive Summary
Raisin, a global fintech company providing an online platform for savings and investment products on multiple markets. A Harness customer for over five years, Raisin has undergone a fundamental change in how they ship software.
Transitioning from a centralized DevOps model reliant on manual Jenkins deployments to a modern platform engineering approach, Raisin achieved complete team autonomy and dramatically scaled its deployment frequency to support the innovation the business needed.
With Harness, Raisin maintains rigorous audit controls at speed, empowering its engineering teams to safely deploy new features and rapidly recover from incidents.
The Challenge: A Delivery Model That Couldn’t Scale With the Business
Raisin had ambitious growth plans that required its engineering organization to scale rapidly. But its software delivery model was becoming a fundamental constraint: adding more teams and developers did not translate into greater delivery capacity.
A centralized DevOps team managed releases using Jenkins and custom scripts. Deployments were coordinated events that took place roughly every two weeks, requiring dedicated release teams and QA involvement. As the organization grew, that model created an increasingly severe bottleneck: more teams led to more changes competing to move through the same centralized release process.
The operational burden was significant as well. Although Jenkins had initially been chosen as a "free" open-source solution, Raisin's DevOps team was spending roughly 80% of its time on administrative toil. Jenkins was fragile, scripts were sensitive, and onboarding a new application took roughly two days.
As a financial services company undergoing multiple audits each year, Raisin couldn't simply remove controls to move faster. It needed a delivery model that could scale organizationally while preserving strict authorization, segregation of duties, and auditability.
The Solution: Platform Engineering and Compliance at Scale
To solve these bottlenecks, Raisin strategically retired three Jenkins instances and fully embraced a decentralized platform engineering model using Code Deployments (CD) in the Harness Software Delivery Agent.
Team Autonomy and Golden Templates
Raisin shifted deployment responsibility directly to individual development teams. Today, every domain can move freely, with teams building, owning, deploying, and operating their own services.
Using a configuration-as-code approach, Raisin created standardized "golden templates". These pipelines are defined once in code and automatically propagate across all workloads. If a change is needed (such as adding a verification stage) it is edited in one code location and automatically applies to all pipelines.
Auditable Controls and Security Posture
For Raisin's partners and customers, security and compliance equal trust. Through Harness's flexible role-based access control (RBAC), Okta integration, and built-in approval flows, Raisin automated key compliance controls.
Nobody has permission to execute changes directly in the UI; everything is managed via code repositories, providing full audit trails. As a result, auditors rarely have questions during regulatory reviews. Furthermore, Raisin maintains their security posture by running Harness delegates within their own infrastructure rather than relying on cloud-hosted execution, ensuring tight control without bottlenecks.
The Results: Transformation and Business Impact
By partnering with Harness to remove the centralized release bottleneck, Raisin created a delivery model that scales with its growing engineering organization.
“Before, every deployment required cross-team coordination. Now, with Harness, engineers have a golden path: required controls and best practices are built directly into the platform, empowering teams to deploy independently and when they’re ready." - Ivan Babenko, Director of Engineering at Raisin
- Accelerated Deployment Frequency: Raisin increased deployment frequency by approximately 400x compared with its Jenkins-based process.
- Previously, releases were coordinated into roughly two shared deployment events per month. With Harness, individual teams deploy independently and on demand. Counting those team-level deployments across the organization, Raisin now performs roughly 400 times as many deployments for a total of 800 monthly deploys.
- Faster Feature Delivery: Decentralized operations enable over 30 teams to deliver value to customers independently and as soon as they are ready.
- Rapid Incident Recovery: Automated rollbacks execute in a few minutes, and have replaced a hotfix process that used to take hours.
“Harness transformed our delivery process into a governed, autonomous engine that effortlessly scales across our product lines. We gained the control and speed our business demands, and our engineers actually enjoy shipping software again.” - Dragomir Prodanov, CTO at Raisin
The Next Frontier: Expanding with AI and Feature Management
As AI-driven code generation increases developer output, Raisin expects its deployment volume to skyrocket. Because their pipelines are entirely configuration-as-code, AI can help create them, while the Harness delegate architecture effortlessly scales to handle increased load.
Raisin is also actively exploring additional Harness capabilities to empower its developers further:
- Model Context Protocol (MCP): To accelerate triage of broken pipelines without taking developers out of their coding environments, Raisin is exploring the Harness MCP to bring SDLC context to Cursor.
- Feature Management and Experimentation (FME): To unify application delivery and feature-level runtime configuration, Raisin is exploring Harness for feature management and experimentation.
