Mobile Developer Productivity Engineering at Europe's Largest Online Fashion Store -Hakkim Alavudeen

Learn how Europe's largest fashion retailer optimizes mobile developer productivity through build engineering, modularization, metrics, and developer experience improvements.

Key takeaways
  • Developer productivity is heavily impacted by infrastructure friction, with slow builds being one of the biggest pain points - waiting for builds is analogous to waiting at traffic lights

  • Modularization of large monolithic codebases can improve build times and developer experience, but requires careful management to avoid module explosion (e.g. Uber manages 12,000 modules)

  • Key metrics for measuring developer productivity include:

    • Deployment frequency
    • Lead time for changes
    • Mean time to restore (MTTR)
    • Change failure rate
    • Developer experience (DevEx) metrics
  • Companies are investing in build engineering optimizations through:

    • Sophisticated build tooling (e.g. Gradle Enterprise)
    • Test parallelization
    • Build caching
    • Cloud/remote IDEs
  • Focus state and uninterrupted coding time significantly correlate with developer productivity - tools like focus bots can help protect developer flow

  • Technical debt management is critical but underinvested:

    • 40% of developer time is spent managing tech debt
    • Only 10% of managers actively invest in tech debt reduction
    • Need better transparency into tech debt impact
  • Mobile development faces unique challenges:

    • Multiple form factors and OS versions
    • Complex release processes through app stores
    • Many teams working on single binary
    • Growing importance with 4x app visit growth vs web
  • Successful developer productivity requires balancing:

    • Infrastructure and tooling
    • Process efficiency
    • Developer experience
    • Business impact
    • Sustainable practices
  • Companies investing in developer productivity gain competitive advantages through faster innovation cycles and better engineer retention

  • Measurement should focus on enabling improvement rather than policing developers - when metrics become targets they cease being good measures