Articles from Source: Spotify-Engineering

At Spotify, data challenges were previously handled by reaching out to experts, but the demand for insights outpaced capacity. To address this, they developed an AI data assistant that efficiently queries over 70,000 datasets, providing reliable answers in seconds. Since August 2025, it's been used by over 2,100 employees across various fields. The assistant ensures trustworthiness through context and ownership, utilizing clusters tied to specific domains and expert teams. #Spotify...

At Spotify, coding is no longer the bottleneck. Niklas Gustavsson, Chief Architect, shared insights on how Spotify is enhancing developer experience through internal platforms and AI. The adoption of AI coding tools surged after the Opus 4.5 release, with over 99% of engineers using them weekly. 🚀 This led to a 76% increase in pull request frequency. Spotify's innovative Fleet Management system has successfully automated over 2.5 million maintenance PRs, allowing developers to focus more on...

Discover how LLM evals are transforming A/B testing at Spotify! 📊 These automated judges assess relevance, coherence, and quality, improving the success rate of experiments. Only 12% of tests yield positive results, but 64% provide valuable insights. LLM evals help identify promising candidates before testing, ensuring experiments focus on driving business outcomes. They generate hypotheses and verify improvements, but experiments are still needed to confirm user responses. Learn more about...

🚀 Spotify continues its journey with background coding agents, known as "Honk," to streamline dataset migrations. In a recent case study, a team utilized Honk alongside Backstage and Fleet Management to simplify the migration of 1,800 direct downstream data pipelines. This automation saved an estimated 10 engineering weeks, demonstrating the power of these tools in managing complex data landscapes. Understanding dataset lineage and utilizing Backstage's features were crucial in this process....

🚀 AI agents are reshaping software development. On March 30, Spotify and Anthropic hosted a discussion at Spotify's London HQ to explore this shift. Key insights included the impact of agentic-first development on daily workflows and the introduction of Honk, Spotify's coding agent. The conversation highlighted the need for context management and the importance of accountability as agents take on more coding tasks. Catch the full chat for deeper insights! 🎤👩‍💻 #AI #SoftwareDevelopment...

🎶 Exciting advancements in Spotify's 2025 Wrapped highlights! This year, Spotify aims to tell personalized stories from your listening history, identifying up to five remarkable days based on your unique music trends. Using advanced algorithms, they capture key moments like your biggest music listening day or most unusual listening day. The process involved generating 1.4 billion reports, ensuring each narrative is data-driven and emotionally resonant. #SpotifyWrapped #MusicTech...

🚀 Spotify's latest article discusses a new multi-agent architecture aimed at streamlining its advertising processes. The focus is on addressing structural inefficiencies in the ads business, where different buying channels operate with their own workflows and decision-making logic. This approach aims to create a unified decision layer, reducing redundancy and ensuring consistent behavior across platforms like Spotify Ads Manager, Salesforce, and Slack. By treating campaign management as...

🚀 Dive into the behind-the-scenes of Spotify's release process in "How We Release the Spotify App: A Look Under the Hood (Part 2)." This article explores the tooling that supports release management. It discusses the transition from using Jira to the new Release Manager Dashboard, designed to streamline information and reduce context switching for managers. The dashboard provides track-specific data, including release state, blocking bugs, and quality metrics. This modernization aims to...

🌟 Exciting news from Spotify! The 2025 FOSS Fund has awarded monetary support to three vital open source projects: 1️⃣ **FFmpeg** - €30,000 A cornerstone for multimedia processing, FFmpeg is essential in streaming and media applications, supporting projects like YouTube and Netflix. 2️⃣ **Mock Service Worker (MSW)** - €15,000 Created by Artem Zakharchenko, MSW simplifies API mocking, enhancing development workflows. 3️⃣ **Xiph.Org Foundation** - Funding details not specified. These projects...

At Spotify, personalization enhances user experience by tailoring content to individual preferences. This is achieved through advanced models that analyze user characteristics and behaviors. 🎧 Experimentation complements personalization by testing and improving these systems. By using a separate tech stack for each, Spotify can optimize both areas effectively. Learn more about the rationale behind this separation and its benefits. #Personalization #Experimentation #SpotifyEngineering...

In Part 2 of our series on context engineering for background coding agents, we delve into effective migration prompts and necessary tools. Spotify has developed a coding agent to enhance our Fleet Management system, capable of editing code, running builds, and automating pull requests. However, guiding this agent effectively has proven challenging. We initially explored open-source options like Goose and Aider but faced difficulties in scaling them for large migrations. This led us to create...

🎶 Spotify has revamped its Shuffle feature to enhance user experience! For years, listeners felt Shuffle wasn't truly random, often hearing the same tracks repeatedly. In response to feedback, Spotify introduced a new system called "Fewer Repeats," which prioritizes freshness while maintaining randomness. This means your favorite songs won't dominate the queue, allowing for a more varied listening experience. Premium users will now find this updated Shuffle as the default option. For those...

Spotify has developed the Confidence platform to enhance product decision-making across numerous teams. The focus has shifted from simply increasing experiment velocity to improving quality and insights through the Experiments with Learning (EwL) metric. This approach emphasizes understanding failures, which helps guide better product decisions. With an aim to foster a culture of experimentation, Spotify has expanded its team involvement from 40 to around 300. This initiative includes...