April 2026 Summaries

Over-the-air (OTA) updates provide a solution for deploying critical fixes directly to users' devices without waiting for app store reviews, leveraging tools such as Expo EAS Update, CodePush, and Shorebird. These updates operate within the boundaries set by Apple and Google, which differ across platforms, with Apple focusing on not altering the app's core purpose or security features and Google allowing interpreted code like JavaScript to bypass direct native code modifications. While Apple's guidelines, particularly Section 3.3(b) of the Developer Program License Agreement, permit OTA updates if they don't transform the app or compromise security, Google's policy explicitly permits JavaScript running in a virtual machine, provided that the app's main functionality and security measures are not compromised. React Native apps, which separate the JavaScript bundle from the native binary, are typically compliant with these guidelines, as long as they adhere to certain conditions such as not introducing new features or altering the app’s primary purpose on iOS, while being more flexible on Android. App developers are advised to conduct checks on native code, functional scope, payments, policy compliance, and rollback strategies before each OTA release to ensure compliance with store policies, and to periodically review these policies for any updates.

Running integration tests for iOS projects presents unique challenges because iOS apps must be built on Apple hardware, while backend servers typically operate on non-Apple systems like Linux. To address this, Bitrise developed a solution that allows integration tests to run effectively by utilizing a combination of workflows and tools within a single pipeline. The setup involves running a Linux workflow to build and operate the server in a Docker container, multiple macOS workflows to build and test the iOS app, and a gatekeeper workflow to manage the testing process. A tunnel, using services like ngrok or Cloudflare Tunnel, bridges the Linux and macOS environments, allowing the iOS app to interact with real server responses instead of relying on staging environments. This approach is particularly beneficial for monorepos, enabling integration tests to be triggered by changes in backend code without the complications of staging environment latencies and state management. However, the reliability of the tunnel depends on the performance of the free-tier services, and high-concurrency builds may require a paid plan for consistent uptime. Bitrise has provided a sample repository to help teams implement this configuration, and further details can be found in the Bitrise Pipelines documentation.

Bitrise has developed two tools, the MCP Server and the using-bitrise-ci agent skill, to enhance the interaction between AI agents and CI/CD pipelines by addressing both access and knowledge gaps. The MCP Server acts as the access layer, connecting agents to the Bitrise API to perform actions like triggering builds and managing projects, while the agent skill serves as the knowledge layer, embedding best practices and configuration conventions into the agent's context. Together, these tools allow an AI agent to not only execute tasks but also understand and apply the appropriate conventions and practices, reducing errors and inefficiencies in the CI/CD process. This dual approach ensures that agents can both propose and validate configurations that adhere to established standards, offering a robust solution to common issues found when using outdated or incorrect configurations. The tools are open-source and designed to work across various platforms, providing structured, secure access and informed decision-making capabilities, which streamline workflow integration and improve code review processes.

Bitrise has implemented a comprehensive competency framework to enhance clarity, transparency, and fairness in career progression and performance assessment across the organization. Initially developed for the Engineering team, the framework includes specific skill, knowledge, and behavior expectations, as well as salary ranges, titles, and job descriptions to support structured career conversations, consistent promotion decisions, and transparent salary discussions. The initiative, motivated by Bitrise's core values of urgency, openness, collaboration, and care, involved individual consultations with engineers to identify gaps and ensure the framework aligned with real-world examples. This collaborative approach resulted in a grounded and credible framework that has since been expanded company-wide, allowing various departments to adapt it to their unique needs while maintaining consistency with the overarching structure. Through annual reviews, the framework remains a living document that evolves with industry changes, ensuring its ongoing relevance and effectiveness.

Bitrise Build Hub offers a specialized mobile CI/CD infrastructure that integrates seamlessly with GitHub Actions, addressing limitations associated with GitHub's Mac runners such as limited hardware and outdated Xcode versions. Build Hub's infrastructure, designed specifically for mobile development, features M4 Pro Apple Silicon machines with 54 GB unified memory, 99.9% uptime SLA, and rapid Xcode updates, ensuring improved reliability and performance. The platform is vertically integrated, owning and managing its hardware and data centers, which allows for efficient resource allocation and reduced environment drift by using ephemeral VMs. This infrastructure supports faster and more reliable builds for both iOS and Android, with substantial improvements in speed and cache efficiency, and includes tools for automated release management to streamline the path from build to production. Bitrise provides tailored machine configurations and boasts a robust caching system that minimizes build times, enhancing overall efficiency and reliability for mobile development teams.

Cirrus CI will cease operations on June 1, 2026, prompting users to migrate their workflows, particularly those using .cirrus.yml files, within eight weeks. Bitrise Build Hub emerges as a recommended alternative, offering enhanced performance with 54% faster Mac builds and 31% faster Android builds, supported by industry-leading machine specifications. It provides a vertically integrated DevOps platform that includes Xcode, Gradle, and Bazel caching, along with mobile release management automation, which significantly reduces build times. ForeFlight's experience with GitHub runners, which led to memory crashes, underscores the reliability and pricing advantages of Build Hub. Additionally, Bitrise's platform supports AI-generated code at production scale, offering comprehensive mobile DevOps solutions such as automated TestFlight delivery, phased Google Play releases, and performance insights. With Cirrus Labs shifting focus to OpenAI's Agent Infrastructure team, Bitrise Build Hub presents itself as a viable long-term solution for both human and AI-driven development workflows.

Bitrise has developed a sandboxed AI agent that enhances its AI strategy by operating on virtual machines within the Bitrise Build Hub, offering greater access to customer code and developer tools compared to its previous AI Assistant. This new agent is designed to run autonomously without requiring engineer intervention and uses pre-defined tool usage lists instead of interactive approvals. The AI agent's framework is provider-agnostic, allowing it to use different LLM providers based on configuration, and it supports prompt caching to prevent performance degradation as context grows. The team encountered a prompt caching issue that skewed success rates during testing, which they addressed by introducing a cache-busting flag. Additionally, session persistence has been implemented to maintain context across queries, and a basic context summarization mechanism is in place to manage the context window limit. The blog post also highlights the introduction of features to monitor agent runtime, token usage, and integrate logging, and mentions the open-sourcing of a simplified version of the core logic for further exploration and development. The final post in the series will discuss scaling AI feature development across Bitrise's R&D organization.