F# in the real world
summary
F# powers backend services for social games serving 1 million DAU and handling 250 million daily requests using AWS (S3, DynamoDB) plus MailboxProcessor actors to keep player state in-memory. A stateful actor design yields ~5x throughput and reduces latency by ~60%, cutting operational cost by ~90%. Domain modeling uses union types, units of measure and type providers; small DSLs for CloudWatch and DynamoDB and genetic algorithms automate balancing and ship features far faster than prior Java implementations.
AI-written summary · about
chapters (8)
- 0:00Introduction and domain modeling
- 5:09Domain modeling with types
- 10:56Actor model for state
- 17:26Asynchronous workflows
- 25:16Event-driven architecture
- 32:09Automating game design
- 34:55Building custom DSLs
- 44:33Type providers and conclusion
related talks
01h 00mF# in the Real World
48mBuilding Reactive Services using Functional Programming
44mLife after business objects. Confessions of an OOP veteran
01h 00mFunctional-first programming with F#
01h 08mWhy Every Day as a Developer Would Be Easier with F#
46mFunctional programming in F# Fad or Fabulous
37mLife After Business Objects: Confessions of an OOP Veteran
01h 07mBuild Life-Saving Apps with Functional DSLs in F#