Coding Is No Longer the Constraint: Scaling Developer Experience to Teams and Agents at Spotify

Summary

Spotify experienced rapid codebase growth alongside widespread adoption of AI coding tools, resulting in a 76% increase in pull request frequency. To manage fleet-wide codebase maintenance and complex refactoring, the company combined its Fleetshift automation platform with Honk, a background coding agent powered by Claude. Honk runs inside Kubernetes pods with access to CI build environments and integrates with Backstage via Model Context Protocol tools to inspect component metadata. Standardized service architectures and active lint guardrails enable the agent to maintain consistency and auto-correct invalid design patterns during execution. By automating routine migrations, Spotify reduced fleet-wide Java backend updates from multi-week cross-team efforts down to three days handled by a single engineer.

Context

Spotify's production codebase expanded seven times faster than engineering headcount, causing developers to spend growing amounts of time on maintenance tasks like dependency upgrades, vulnerability patching, and API migrations. Concurrently, AI tool adoption surged to over 99% of engineers weekly, driving a 76% increase in pull request volume and moving the development bottleneck from writing code to human decision-making and review.

Approach / What changed

Spotify paired its Fleet Management orchestration tool, Fleetshift, with Honk, a background coding agent utilizing Claude via the Agent SDK deployed in Kubernetes pods. Honk runs builds across operating systems in CI to verify changes and can be triggered via Slack or Fleetshift. Spotify leveraged Backstage to expose component catalogs and documentation to agents via MCP and CLI tools, using golden state definitions, Soundcheck self-assessments, and linting systems to provide automated feedback and enforce architectural consistency.

Takeaways

• Standardized technology stacks and consistent architectural design patterns directly improve LLM coding performance compared to fragmented codebases.
• Combining automated fleet orchestration with background coding agents reduced a backend Java migration across services to three days for one engineer.
• Exposing internal developer portal capabilities and linting systems to agents allows them to discover component ownership, verify builds across OS environments, and self-correct invalid patterns.

Related reading

• Spotify · Jun 10, 2026

  Encoding Your Domain Expert: The Context Layer Behind Spotify's Data Assistant

  Spotify developed an AI data assistant, Vedder, to scale access to over 70,000 internal datasets without overwhelming domain experts. Because raw warehouse schemas omit critical semantic nuances and overwhelm LLM context windows, engineers implemented a domain-owned context layer organized into clusters. Each cluster contains profiled datasets, expert-curated business documentation, and vetted question-and-SQL pairs that guide a ReAct-based agent. Automated query history extraction proved insufficient on its own, as domain curators accepted only 12.5% of inferred query pairs due to noise. Deployed across Slack, IDEs, and a dedicated web interface, the system maintains reliability through continuous cluster health monitoring that flags schema drift and degraded examples.

  Spotify Engineering

• Spotify · Jul 27, 2026

  Indexing the Data Lake for Online Point Queries