Skip to content
Back

What I Have Learnt as a Frontend Engineer

Four lessons from building and operating production web and mobile products

Four years ago, I thought frontend engineering meant turning designs into polished interfaces. Production taught me to look beyond the component: architecture, data freshness, reliability, delivery, and measurable user outcomes are all part of the job.


The metrics below are project outcomes, not universal benchmarks. What matters is the link between a technical decision and an observable result.


Outcomes at a Glance

  • Architecture: 80% faster page loads, 90% lower search latency, and PageSpeed above 90.
  • Web performance: LCP below 0.5 seconds and CLS below 0.1.
  • Mobile reliability: More than 99.5% crash-free sessions after launch.
  • Delivery: Build and deployment time reduced by more than 50% with SST.

Four Lessons

  1. Architecture changes beat local optimizations. Measure the full path before tuning the React tree.
  2. Caching is a freshness policy. Static content, editorial updates, and mutable data have different lifecycles.
  3. Reliability is user-facing. Dropped frames, memory growth, loading failures, and crashes interrupt the same journey.
  4. The delivery system is part of the product. Shared foundations and safer releases compound across teams.


What “Senior” Means to Me Now

  • Measure the user journey, not only component render time.
  • Match the architecture to the bottleneck, even when it sits outside the frontend repository.
  • Model freshness and ownership instead of adding caches reactively.
  • Operate the product after release with real reliability signals.
  • Improve the path for the next engineer through shared foundations, tests, and safer delivery.


Framework APIs will change. The judgment behind this loop should remain useful.