April 2021 Summaries

Lano, an HR software provider, experienced rapid growth during the COVID-19 pandemic, leading to increased pressure on their engineering infrastructure. The company's Ops team struggled with complex workflows, custom scripts, and high maintenance overhead, causing delays and frustration among developers. A critical integration blocked staging for two weeks, highlighting the need for a more scalable CI/CD setup. To address this, Lano implemented an Internal Developer Platform (IDP) built using Humanitec, which provides a lightweight API layer, enabling true developer self-service and automating workflows. The IDP has improved DevEx, reduced MTTR by 40%, and allowed the Ops team to support more developers without issues.

Forte, a digital freight company, initially used Heroku due to its ease of use and the value placed on engineers' time. However, as their team grew and the application became more complex, they faced issues with autonomy and configuration management. The company realized they needed to switch to Kubernetes to address these challenges and planned a gradual migration process. They introduced Dockerization, standardized tools and patterns, and hired an SRE team to manage configuration management and secrets. Forto's key learnings from the migration include understanding the business domain and relating infrastructure choices to company goals, and leveraging commercial tools such as Humanitec for internal developer platforms. The company aims to expand their Kubernetes setup to multiple regions, implement audited escalations for engineers, and move towards a passwordless architecture.

The emerging category of Internal Developer Platforms (IDPs) is revolutionizing the way teams work, providing self-service capabilities that enable developers to build and deploy applications quickly and efficiently. IDPs are built on top of platform engineering principles and consist of six key areas of functionality: infrastructure orchestration, application configuration management, role-based access control, deployment management, environment management, and observability. These components work together to provide a seamless developer experience, allowing developers to spin up fully provisioned environments on demand, manage these environments at scale, and surface container logs and deployment errors. By automating transactional tasks and providing a clear audit trail of every single change, IDPs transform the way teams work, enabling a true "you build it, you run it" mindset with an entirely different ownership culture.

Humanitec, a platform used to build Internal Developer Platforms, has announced the general availability of Persistent Volumes, simplifying the process of using volumes in Kubernetes for developers and ops teams alike by taking care of Persistent Volume Claims. The volume resource is defined similarly to provisioning other resources such as databases or S3 buckets, allowing easy management of different environments with varying volumes and sharing of volumes across environments. This enables developers to focus on writing code without worrying about volume dependencies, while ops teams can ensure data stays in the intended environment.

Internal Platform teams are becoming increasingly important for large organizations to ensure end-to-end developer self-service and streamline development and deployment processes. These teams build internal workflows and tooling, such as Internal Developer Platforms (IDPs), to shield application developers from infrastructure complexities. By enabling developer self-service across the organization, platform teams create a smooth experience for everyone involved. Top-tier engineering organizations have found that building an internal platform team is crucial for scaling rapidly while maintaining reliable and maintainable systems. These teams focus on reducing complexity and cognitive load for app developers, treating their platform as a product, optimizing iteration speed, solving common problems, and providing glue to bring together various tools. By doing so, they can help organizations scale considerably faster and more efficiently than with a less standardized approach.