August 2025 Summaries

Feature Health Beta, introduced by Flagsmith, enhances the observability of feature flags by allowing users to monitor metrics directly within the Flagsmith dashboard using Grafana as the first provider. This integration connects feature changes to operational health data, enabling users to quickly identify and address issues such as performance degradation or errors related to specific feature flags. By treating feature flags as observable systems, Feature Health provides immediate feedback on rollout performance, facilitates faster responses, and makes releases safer by offering insights into the impact of feature toggles, thereby reducing the risk of enabling advanced flags in production.

In an era of rapid technological and regulatory changes, organizations are increasingly expected to accelerate their software delivery while maintaining quality, security, and compliance. Contrary to the traditional notion that speed compromises safety, research by Dr. Nicole Forsgren and her DORA team reveals that adopting certain practices can enhance both pace and stability. High-performing organizations achieve this through Continuous Delivery techniques that expedite feedback loops and reduce manual work, thereby allowing them to deploy code more frequently without compromising quality. A key practice in this approach is feature flagging, which separates deployment from release, enabling incremental code deployment that remains dormant until fully ready. This method not only supports frequent and safe deployments but also facilitates practices like A/B testing, canary releasing, and dark launching. While feature flags can impose a technical burden if unmanaged, strategic adoption can significantly enhance agility and speed in software development processes, a transformation observed over recent years as feature flagging has evolved from a niche concept to a mainstream practice in mature engineering organizations.

As the co-founder and CTO of BibliU, an EdTech SaaS company, the author reflects on the decision to build an in-house feature flag system rather than purchasing an off-the-shelf solution, due to strict data control agreements and a belief in their unique needs. The process of building internally was chosen despite the challenges it posed, such as cross-team development coordination and the need for extensive documentation and maintenance. While the in-house system offered customizability and control, it also required careful monitoring and had higher operational costs, leading to considerations of different approaches for future projects like HowdyGo. The author emphasizes the importance of evaluating whether a feature is a core competency, the criticality of the system, costs, scalability, and regulatory requirements when deciding between building in-house or opting for third-party tools, applying these lessons to make more informed decisions in their new venture, HowdyGo.

Deciding whether to build or buy feature flagging software involves weighing various factors and understanding where your team's needs lie on a continuum from in-house development to third-party solutions. Initially, simple and specific flagging requirements may be effectively managed internally, especially for small teams or niche use cases. However, as flagging needs become more complex—such as when scaling up, transitioning to microservices, or managing multiple languages and teams—purchasing a feature flagging tool can streamline processes and free up engineering resources. Key considerations include data protection, feature management complexity, cost-effectiveness, scalability, and the risk of vendor lock-in. Open-source solutions like Flagsmith offer flexibility and transparency, allowing teams to switch providers if needed. Ultimately, the choice depends on the company's specific requirements, resources, and long-term objectives.

In today's dynamic software development environment, feature flags have emerged as a vital tool for managing feature releases, providing developers with the ability to toggle features on or off without redeploying code, thereby enhancing flexibility and reducing risk. The text explores the implementation of feature flags in a Java Spring Boot application using Flagsmith, a feature flag management tool, demonstrated through a banking app. By integrating Flagsmith, developers can efficiently manage the release of new features, conduct A/B testing, and tailor functionality to specific user segments. The guide also delves into advanced usage, such as storing and controlling feature values and optimizing response latency through local evaluation mode. By adopting Flagsmith, development teams can streamline their deployment process, improve user experience, and respond swiftly to issues, underscoring the significance of feature flags in modern application development.

Feature flags are a critical tool in modern software development, offering a way to manage risk and improve the deployment process by decoupling the deployment of code from its release. They allow teams to implement various deployment strategies, such as big bang, blue-green, canary, A/B testing, progressive delivery, rolling, shadow, and recreate deployments, each serving different needs and risk levels. By using feature flags, developers can gradually introduce new features to specific user segments, test their impact, and quickly disable them if issues arise. This approach not only supports continuous delivery by enabling faster feedback loops and real-time monitoring but also facilitates smoother transitions in modernization efforts, such as moving from monolithic architectures to microservices. Feature flags are particularly valuable in industries where data sensitivity is high, allowing teams to innovate while minimizing disruption and risk.

Feature flags, also known as feature toggles, are a crucial tool in software development that allow developers to enable or disable features dynamically without deploying new code, offering flexibility and control over feature releases. In a React.js application, feature flags can be used with Flagsmith, a platform for managing feature flags, to safely deploy a feature such as displaying article reading times, which can be toggled on or off without requiring redeployment. This tutorial guides users through setting up a React.js application to list articles from the Dev.to API, integrating Flagsmith to manage feature flags, and deploying the application using Surge.sh. The key advantage of feature flags is the ability to mitigate risk by quickly disabling problematic features, enhancing the safety and efficiency of software releases. This concept can be extended to various frontend applications, enabling safe and flexible feature deployment across different platforms.

Observability in software engineering has evolved from simple Application Performance Monitoring (APM) of static infrastructure to a complex, integrated approach addressing the dynamic nature of modern cloud-native environments. Unlike the past when applications were monitored post-production, modern observability requires a "Shift-Left" approach, involving developers in defining and implementing observability as part of the software development lifecycle. This involves using standards like OpenTelemetry to emit traces and metrics, ensuring that observability data is collected and analyzed at every stage of the software delivery pipeline. As infrastructure becomes more complex with services like Kubernetes and serverless offerings, and as progressive delivery techniques like feature flags become prevalent, there is an increased need for observability to ensure system reliability, security, and resilience. Feature flags, for example, must be observable to assess their impact on user experience and infrastructure, necessitating robust observability solutions that provide real-time insights and automated decision-making capabilities. Organizations that integrate modern observability practices into their development processes are positioned to deliver more reliable, secure, and innovative solutions.

Flagsmith has been actively enhancing its open-source feature flag management software to improve release workflows and address customer needs, particularly focusing on security and enterprise-level management. Recent updates include new integrations with GitHub and Grafana, which enhance visibility and maintenance of feature flags by linking them to issues and pull requests, and allow changes to flags to be reflected in Grafana metrics. Improvements have been made to the local evaluation engine, enabling identity overrides, and an updated permissions system now allows for more granular control through roles. Additional features such as custom fields and quality of life updates aim to streamline navigation and flag management. Looking ahead, Flagsmith plans to introduce feature versioning, stale flag identification, transient traits and identities, and enhanced observability integrations to provide better control and insight into feature releases.

Flagsmith's journey to optimize its SaaS infrastructure over time reveals a series of strategic decisions aimed at balancing cost, scalability, and reliability. Initially launched as a side project, the platform utilized cost-effective solutions like Hetzner and dokku, with minimal redundancy and scalability, primarily because it had zero customers. As usage grew, notably due to a large UK lottery operator, Flagsmith transitioned to AWS Elastic Beanstalk, which provided necessary features such as auto-scaling and managed databases, despite some deployment challenges. This setup allowed the company to handle a significantly increased load, though database connections eventually became a bottleneck. The next evolution involved moving to AWS ECS and Fargate, leveraging Docker and containerization to align more closely with enterprise standards, resulting in improved performance and reduced anxiety during deployments. With costs stabilizing and infrastructure solidified, Flagsmith plans to further enhance its system by migrating SDK traffic to a globally distributed Edge API, promising to maintain its commitment to efficient and scalable infrastructure.

Flagsmith's strategic use of feature flags enabled a successful and smooth rollout of a pricing update, crucial for implementing automated alerting and billing for API overages in its SaaS environment. By carefully planning and segmenting the rollout, Flagsmith avoided customer backlash typically associated with pricing changes, especially given that such changes were historically challenging due to the lack of automatic monitoring for API overages. Initially testing functionality in a staging environment, and later incrementally deploying it to both free and paid customer segments, allowed the company to catch and correct errors, such as an undercharging issue, without disrupting the entire customer base. This approach provided the flexibility to enable, adjust, or disable features as needed, ensuring accurate billing and maintaining customer satisfaction while also safeguarding the platform from misuse.

Navigating authorisation management presents significant challenges for companies, particularly as they face dynamic requirements and the limitations of legacy systems with hard-coded logic. For regulated industries, maintaining compliance with standards like ISO27001 or SOC2 is complicated by fragmented audit logs that hinder efficient security incident responses. Transitioning to modern authorisation involves decoupling authorisation logic from application code, centralising it in audited and version-controlled policies via a Policy Decision Point (PDP), which allows for independent evolution and consistent audit trails. This approach not only enhances technical efficiency and compliance but also empowers non-expert stakeholders to understand and modify permissions without extensive coding knowledge. However, adopting decoupled authorisation requires careful evaluation of potential risks, such as vendor lock-in and added latency, ensuring that the chosen solution is scalable, performant, and integrates seamlessly with existing infrastructure. By addressing these challenges, businesses can improve their security, scalability, and transparency while maintaining a robust and adaptable authorisation framework.

Feature flags enhance developer confidence and release velocity by allowing code to be pushed to production without being blocked by dependencies, thus improving daily workflows and reducing impediments. By wrapping features in flags, developers can enable them when dependencies are ready, avoiding blockages in builds, code, and releases. This approach also minimizes the occurrence of messy branch situations and large pull requests, as production is updated more frequently, leading to fewer long-lived feature branches and a more stable release process. Despite the perception that rapid releases may decrease stability, the opposite is often true, as smaller, iterative releases reduce the chance of errors. While feature flags positively impact quantitative metrics like reduced pull request times, lower incident rates, fewer merge conflicts, higher rollout frequency, and more successful rollbacks, the primary benefit remains the qualitative improvement in developer confidence, leading to more regular and confident code releases.

In 2024, a flawed software configuration released by CrowdStrike caused a global outage affecting banking systems, airlines, and healthcare providers, resulting in significant financial and reputational damage. This incident highlighted the importance of implementing feature flags to mitigate risks associated with software updates. Feature flags allow companies to roll back changes quickly or conduct canary releases, where updates are initially distributed to a small percentage of users to monitor performance before a broader rollout. Flagsmith offers solutions for integrating feature flags, enabling organizations to manage software releases more effectively and prevent widespread disruptions.

In the evolving banking sector, managing legacy infrastructure presents challenges such as scalability issues, security vulnerabilities, and inefficient resource use, prompting banks to transition to modern platform solutions. These platforms promise regulatory compliance and operational streamlining, allowing banks to integrate the latest technologies without the complexities of on-premises management. While modern platforms facilitate smoother integration with legacy systems through APIs and offer advantages in compliance and scalability, transitioning requires significant cultural shifts within traditionally siloed banking organizations. Adopting a GitOps-based approach, which uses Git as a source of truth for Infrastructure-as-Code, can enhance collaboration, compliance, and security, although it demands technical expertise and organizational adaptation. Ultimately, platform engineering offers banks a holistic solution to modernize infrastructure, improve efficiency, and foster innovation, despite the need for gradual cultural change and change management strategies.

Transitioning from managing basic booleans in-house to adopting feature flag management tools can be challenging, but with the right approach, teams can effectively utilize these tools to improve their release processes. The process involves identifying the challenges and needs that necessitate a feature flag tool, such as managing coordinated releases and facilitating A/B testing. Once identified, teams should focus on configuring the tool, establishing best practice guidelines, and setting up testing and monitoring processes. Onboarding teams to the new system requires thorough training and solid documentation to ensure understanding and alignment. Shifting the cultural mindset towards modern release practices, like continuous deployment, is crucial for maximizing the tool's benefits. Continuous monitoring and improvement of the developer and product experience are essential for success, with the ultimate goal of enhancing confidence in feature releases.

Komerční Banka (KB), a prominent Czech Republic bank within the Société Générale group, has embarked on a significant modernization initiative featuring an Agile 2.0 model and microservices architecture to enhance flexibility and reduce time-to-market. Transitioning from a monolithic internet and mobile banking system to one with microservices, KB recognized the need for agility, adopting trunk-based development and feature flags, facilitated by Flagsmith, to manage unfinished code and pilot new features in production. This strategic transition involved a comprehensive overhaul of internal processes, organizational structures, and the introduction of technologies like Kubernetes and OpenStack, supported by leadership buy-in and external consultants. Central to this transformation was the establishment of a platform team that equips business teams with the necessary tools, frameworks, and infrastructure, ensuring streamlined and effective development workflows. The adoption of feature flags required careful governance to manage access, usage, and the integration of new features, with particular attention to user experience and communication in complex cases such as implementing kill switches.

Observability-Driven Development (ODD) is an innovative approach that emphasizes real-time monitoring and understanding of code performance, behavior, and impact, enabling developers to adapt features without redeployment. Central to ODD are feature flags, which allow developers to toggle features on or off at runtime, facilitating conditional control, gradual rollouts, and A/B testing. Integrating feature flags with observability tools enhances the ability to track feature usage, perform diagnostics, and gather analytics, contributing to continuous improvement and more informed development decisions. Implementing ODD involves setting clear objectives, selecting appropriate observability tools, instrumenting key metrics, employing distributed tracing, and creating meaningful alerts to ensure a resilient development process. This methodology not only improves feature deployment precision but also fosters a collaborative and adaptable development environment that aligns with user expectations and business goals.

Inflow, a mobile app designed to aid individuals in managing ADHD, has successfully integrated A/B and multivariate testing into its development process using Flagsmith and Mixpanel to optimize user experience and product understanding. The app uses React Native for its frontend and Node.js for its API, with Mixpanel already in place to track user behavior. By implementing a 50/50 A/B split test through Flagsmith, Inflow can effectively bucket users and analyze the results of various experiments, such as adding animations or instructional text to improve user engagement. The app also employs simple role-based access control to manage content access for creators. Through these tests, Inflow has been able to gather actionable insights, such as the impact of animations on user engagement, and plans to continue experimenting with other user flows, including pricing models and onboarding videos, to further enhance their app's effectiveness for those managing ADHD.

The author shares their optimized Mac setup, emphasizing tools that enhance productivity and workflow efficiency. Key components include a clipboard manager, such as the one in Alfred, which boosts productivity by storing and searching a long history of copied items, and Hammerspoon, a versatile window manager configured with keyboard shortcuts that streamline window management across different work contexts. Additional tools like Zappy for screenshots, Alfred for launching applications and creating shortcuts, and Infinite History for storing and searching command-line inputs further enhance the setup. The author also highlights their preference for specific applications like Table Plus for database management, Paw for RESTful APIs, Sublime Merge for Git, Mimestream as an efficient Gmail client, and self-hosted Bitwarden for password management. Despite small annoyances like the iVANKY dock issue with Zoom, the author expresses satisfaction with their current setup, indicating its effectiveness in maintaining productivity.

Grafana is widely recognized as the standard observability tool used by engineering teams globally for monitoring traffic patterns, user activity, and other metrics in production environments. The integration of Flagsmith with Grafana allows users to incorporate feature flag change events as annotations within Grafana dashboards, offering insights into the impact of these changes on applications. This integration is particularly useful for visualizing app performance and metrics when feature flags are toggled, such as in a Next.js website setup. Setting up this integration involves creating a service account and token in Grafana, configuring the Flagsmith dashboard, and ensuring public accessibility of the Grafana dashboard. Once integrated, users can filter annotations by various criteria, such as user email or environment, to track when specific feature flags are enabled or disabled, thereby enhancing visibility and enabling more informed decision-making without the need for separate monitoring dashboards.

Feature flags are initially easy for development teams to manage independently, often as simple booleans in the database or code configuration, but as projects grow, in-house management becomes unsustainable, necessitating dedicated feature flag software. Key indicators for transitioning to such tools include the need for role-based access control, auditing, and efficient testing in production environments; these tools facilitate real-time testing and precise control over feature releases, addressing issues promptly and minimizing user impact. In complex setups, like multi-team environments or mobile application development, feature flags allow for coordinated releases and graceful degradation, ensuring a smooth user experience even when bugs arise. Additionally, feature flag management systems empower product owners by providing a user-friendly interface to toggle features without engineering intervention, enhancing visibility and control over feature performance. To ensure a successful transition, it is essential to assess deployment options, data sensitivity, integration capabilities, and migration support, while also considering team training and the gradual rollout of new tools. Moving to dedicated feature flag software can significantly streamline deployment processes, reduce risks, and align with agile development practices.

The text explores the pros and cons of bootstrapping versus venture capital (VC) funding for startups, emphasizing that the suitability of each path depends on various factors such as the capital intensity of the business, the founder's growth ambitions, and the company's long-term goals. It highlights that bootstrapping can be advantageous for less capital-intensive businesses like software companies, allowing them to grow at their own pace and focus on niche markets without the pressure of rapid scalability. In contrast, VC funding is often necessary for capital-intensive ventures and those seeking quick market dominance, as it provides substantial financial resources and strategic partnerships. However, the text cautions that VC-backed companies face pressures to achieve high valuations and rapid growth, which can sometimes lead to unsustainable practices. The experience of employees in both scenarios is also discussed, noting that bootstrapped companies may offer more realistic financial gains through dividends or acquisitions, whereas VC-funded firms could promise larger, albeit less certain, payouts. Ultimately, the text underscores that the decision between bootstrapping and seeking VC should align with the founder's personal and business objectives, acknowledging that both paths have inherent challenges and risks.

In 2001, a London-based software agency was founded, which, after years of collaboration with diverse engineering teams, identified challenges in software shipping and subsequently developed Flagsmith, an internal feature flagging tool. By 2018, the project was open-sourced and launched as a paid SaaS product, leading to a significant revenue increase after its commercial open-source debut. Over the past year, the company has emphasized five key learnings: ensuring flexibility in deployment to broaden market reach, removing growth impediments before scaling, focusing on a clearly defined target customer, offering outstanding customer support, and continuously learning from others. These strategies have helped Flagsmith become a reputable alternative in the feature flagging market, foster a strong community engagement, and significantly grow their user base and revenue.

Flagsmith's recent infrastructure overhaul involved transitioning to a serverless architecture with the launch of its Global Edge API, aimed at enhancing global low latency, failover capabilities, and cost management. The API is divided into two main tasks: serving flags, which requires high uptime and low latency, and handling dashboard requests, which are less latency-sensitive. The new architecture uses AWS Global Accelerator for routing and failover, Lambda for compute, and DynamoDB global tables for data storage. The switch to serverless, including a move to arm/Graviton instances, reduced costs by 30% without performance loss. Lambda compute costs about $3 per million requests, while DynamoDB global tables cost roughly $6 due to the read-heavy nature of the platform. Global Accelerator adds minimal cost at about 20 cents per million requests, and data transfer costs are approximately $1.50 per million requests. Despite higher monthly costs, the serverless setup offers significant scalability and failover advantages, aligning with Flagsmith's priorities and reducing the complexity and risk associated with traditional database scaling.

Firebase, a popular choice for application development due to its user-friendly features, faces criticism from developers seeking remote configuration solutions due to issues such as vendor lock-in, analytics data discrepancies, and security challenges within the Google ecosystem. These drawbacks often lead developers to consider alternatives, such as Flagsmith, which offers a dedicated, open-source platform with robust feature flagging and remote configuration capabilities, suitable for security-conscious organizations. Other alternatives include building bespoke solutions, ConfigCat, AWS AppConfig, Unity, and Azure App Configuration, each offering unique features and pricing models. These alternatives address various limitations of Firebase, such as caching issues, rate limits, and analytics integration, making them viable options for teams needing more flexibility and control over application configurations and feature releases.

Vontobel, a globally active investment firm, leverages feature flags to enhance its software development process, using Flagsmith to improve release confidence and streamline operations. By securely hosting Flagsmith on their infrastructure, Vontobel allows product owners to toggle features independently, freeing up engineering resources and increasing development velocity. This transition from a homegrown solution to Flagsmith has improved visibility and control over feature changes, enabling the firm to manage features across different stages more efficiently. Feature flags are used to hide backend features, manage proof of concepts, and communicate maintenance activities to customers, which aids in maintaining a single code base and avoiding extensive deployments. Vontobel's integration of Flagsmith supports multiple projects within one SDK, and the company values strong support and compliance with GDPR and banking protocols when adopting new tools. The introduction of feature flags has significantly increased Vontobel's development speed and provided peace of mind regarding production releases.

The integration of Flagsmith with Datadog enhances feature toggling capabilities, allowing users to monitor feature flag updates alongside their existing tech stack. This guide provides a tutorial on setting up the integration, which enriches Real User Monitoring (RUM) telemetry with Flagsmith's data for better insight into application performance and user experience. Users can manage feature flags directly from Datadog's interface, streamline workflows, and create monitors for alerting on critical changes. This unified platform approach allows for efficient tracking of feature flags and their impact, aiding in the identification and resolution of performance issues. By leveraging this integration, companies can maintain a comprehensive view of their application's functionality, ensuring smoother operations and more responsive troubleshooting.

Feature flags, also known as feature toggles or switches, provide a method for software teams to separate the deployment of code from the release of new features, allowing for greater flexibility and control over the release process. By using tools like Flagsmith, developers can implement feature flags to enable or disable functionalities without altering the codebase. This approach allows for controlled testing in production environments with a select group of users, minimizing the potential impact of bugs and reducing the blast radius in case of errors. A practical example is given, showing how to use Flagsmith to manage feature flags for displaying ads on a blog, demonstrating the ease of switching features on or off without redeployment. This decoupling empowers business stakeholders to make release decisions independently, enhancing agility and responsiveness in software development.