Elixir for curing product development pains
summary
Elixir powers a prototype that becomes a scaling travel backend by handling unreliable third‑party APIs with lightweight concurrency and fault tolerance, checking roughly 5 airports per origin which generates about 15 requests per query, and letting the team rewrite routing logic into a faster algorithm. The stack uses Phoenix, Ecto and StreamData for property tests, publishes OpenAPI specs to speed front‑end integration, achieves a 10x faster API in one change, and abandons incremental Dialyzer work after Dialyzer produced 138 failures.
AI-written summary · about
chapters (12)
- 0:00Start
- 0:06(An) Elixir for curing product development pains?
- 3:34Let me take you on a journey.
- 7:27Finding a balance between time, scope and quality.
- 22:36Remember?
- 26:37Example #2
- 28:28Example #3
- 32:57Type safety
- 34:07Let's change tactics and talk validation
- 37:59Is Elixir a cure for product development pains?
- 38:37Elixir enabled us to work on the product, not on production
- 39:12Gradual rollout New routing 10x faster API
related talks
44mElixir & Phoenix - Fast, Concurrent and Explicit
38mElixir Adoption at Scale: A Comprehensive Guide
19mElixir – what it is and why is it amazing?
27mHow Elixir is driving Australian product innovation
30mCan Elixir Bring Down Phoenix?
38mBuilding and Testing API Endpoints with Phoenix
41mA Brief History of Elixir
01h 52mPurple Carpet