May 2025 Summaries

Harness Blue-Green Traffic Shifting introduces a new approach to deploying applications that incrementally redirects traffic between existing ("blue") and new ("green") environments, reducing deployment risks and enabling real-time validation. This feature allows for precise control over deployments on AWS ECS, Auto Scaling Groups, and Spot Elastigroup, incorporating built-in approval gates and automated rollback criteria to ensure stability and confidence. Traditional blue-green deployments often expose all users to potential issues in the new environment, but Harness's incremental rollouts mitigate this by progressively adjusting traffic weights, which limits the impact of any undetected defects. The system allows for monitoring key metrics such as latency and error rates, facilitating informed decisions before proceeding with further rollouts. The mechanism supports AWS Application Load Balancer configurations and integrates with existing tooling like CloudWatch and Datadog, enhancing operational insight and enabling quick and controlled rollbacks if needed. This innovative method is poised to transform how deployments are managed, offering a lower-risk alternative to traditional full-traffic shifts.

Harness has introduced the MCP Server, a groundbreaking tool designed to enhance the integration of AI agents within software delivery workflows by using the Model Context Protocol (MCP). This server acts as a translator between AI tools and Harness's platform, enabling secure, structured access to essential developer services without the need for complex custom APIs or fragile code. Built with enterprise-grade security, the MCP Server facilitates seamless interactions between AI agents and existing tools like pipelines, logs, and artifact registries, allowing teams to automate and scale their processes efficiently. This innovation represents a shift towards integrating AI as a core component of the software delivery lifecycle, enabling developers to focus on execution without compromising security or efficiency, and reflects Harness's commitment to intelligent and secure software delivery solutions.

Fidelity Investments successfully migrated over 50,000 state files from Terraform to OpenTofu, demonstrating that such a transition, with the right planning and strategy, is more of an organizational challenge than a technical one. The migration was driven by HashiCorp's shift to BSL licensing, prompting Fidelity, committed to open-source principles, to align with the community-driven OpenTofu under the Linux Foundation. Fidelity's approach involved a six-phase strategy focusing on rigorous proof of concept, lighthouse projects, consensus building, developer enablement, progress tracking, and a default switch, without requiring extensive code rewrites due to OpenTofu's compatibility with Terraform. This process emphasized the importance of organizational factors such as stakeholder engagement and infrastructure adaptation over technical challenges, with Fidelity actively contributing to the OpenTofu community to shape the future of infrastructure automation.

Harness Software Engineering Insights (SEI) is designed to transform scattered engineering metrics into actionable insights by integrating DORA metrics, PR analytics, and business alignment into role-based dashboards tailored for different personas within an organization. This tool enables engineering teams to align their execution with business goals by offering structured rollout frameworks and outcome-driven OKRs, which help in building an insight-driven culture. SEI emphasizes the importance of starting with clear objectives, choosing relevant KPIs, and configuring the platform for actionable insights, all while aligning metrics to specific roles to reduce dashboard fatigue and increase engagement. Real-world case studies highlight SEI's impact across various industries, such as cybersecurity, financial services, and gaming, where it has improved metrics like PR cycle time, deployment frequency, and change failure rates, thereby enhancing engineering efficiency and product delivery. Adeeb Valiulla, who leads the engineering excellence efforts at Harness, advocates for using SEI to operationalize metrics and foster a culture of continuous improvement and accountability, ensuring that engineering organizations can unlock their potential through data-driven strategies.

The Harness Developer Hub (HDH) has achieved a milestone of 10,000 pull requests, highlighting the importance of documentation as an evolving product. By implementing a developer-first, docs-as-code approach, HDH has turned documentation into a collaborative asset that evolves with the product. This model allows contributions from across the organization, reflecting the dynamic nature of the software delivery process. Initially relying on a centralized documentation model, HDH was launched to consolidate learning resources and promote decentralization, enabling direct contributions from product teams and community members. This approach, supported by a Git-based workflow, treats documentation with the same rigor as software development, ensuring it is versioned, validated, and improved continually. The platform has become a vital part of the product experience, enhancing onboarding and adoption, as seen in the success of Harness University. The community-driven contributions, often small yet impactful, have transformed HDH into a platform owned by the community, demonstrating the power of treating documentation as an integral product component.

Microsoft's study, "Time Warp: The Gap Between Developers' Ideal vs Actual Workweeks in an AI-Driven Era," reveals a significant disconnect between how developers currently spend their workweeks and how they ideally would like to spend them. Developers often find themselves bogged down by meetings, communication tools, and administrative tasks, leaving little time for deep, meaningful work such as coding and problem-solving. The study underscores the importance of creating work environments that enable developers to focus on impactful tasks, with AI tools like GitHub Copilot helping to streamline workflows by automating repetitive tasks. To address this gap, it is essential to first understand the current developer experience through qualitative and quantitative metrics, such as DORA metrics, and then take deliberate actions to remove bottlenecks and enhance productivity. The ultimate goal is not just to boost efficiency but to restore a sense of purpose and creativity in developers' work, allowing them to engage in the tasks they find truly fulfilling.

In Q1 2025, Harness introduced significant enhancements to its GitOps framework, focusing on advanced sync controls, rollback support, and improved observability, which aim to bring GitOps pipelines closer to Continuous Delivery (CD) parity. These updates include OpenShift-certified integration, support for the valuesObject field, audit logs for manual changes, API updates for agent management, and Prometheus monitoring for GitOps agents. Additionally, the enhancements feature rollback support in pipeline steps, improved service and environment dashboards, and enriched log viewing for easier debugging and log exploration. The updates are designed to offer users better control, transparency, and scalability, particularly in regulated environments or dynamic infrastructures, with detailed guidance available through the Harness GitOps Documentation and GitHub samples.

Automating environment-specific verification queries is a method to ensure database changes are validated accurately across development, staging, and production environments, despite data discrepancies. By utilizing tools like Liquibase and Harness Database DevOps, teams can streamline workflows, catch issues early, and validate changes consistently. Verification queries act as unit tests for database changes, which are crucial for maintaining data integrity, especially when environments have varying data sets. The solution involves leveraging environment-specific properties, Liquibase changesets, and stored procedures to execute and compare query results. This automated approach reduces manual intervention, allows for faster feedback, and ensures only validated changes are applied, enhancing the developer experience by keeping change scripts environment-agnostic. The process also includes defining properties in a centralized manner and thoroughly testing changes in non-production environments to maintain consistency and minimize errors. Stephen Atwell, who has a diverse background in IT and software development, leads Harness's Database DevOps product, focusing on solutions that improve efficiency and compliance in database management.

Harness' Q1 2025 release introduces significant enhancements across Kubernetes, ECS, Terraform, and GitOps, enhancing precision, governance, and flexibility in deployment processes. Key updates include a native Kubernetes diff step for previewing changes, strict namespace consistency enforcement, and improved traffic routing control in Kubernetes. ECS sees advancements with weighted traffic shifting in Blue Green deployments, while Terraform enhancements allow editable Git details and auto-approval of refresh runs. Artifact management now supports more informative metadata exposure, and governance features include consolidated approval notifications and environment-type-based freeze windows. The platform sees performance upgrades with a protocol buffer update and improved cluster visibility in the Rancher connector. These updates aim to enhance reliability, scalability, and user empowerment, reflecting Harness' commitment to simplifying complex deployment workflows as it continues to innovate into 2025.