Total Rewrite: A Story About F# and Azure Functions in Production

on YouTube ↗11m

" summary

Rewrites an existing C# App Service system to F# and Azure Functions and ships a serverless headliners API that reads geo‑redundant blobs behind nginx+varnish, using blob/table/queue bindings, Azure DevOps and Expecto. Pipeline runs nightly: recommendation importer → catalog fetcher → publisher. May metrics show 5M blob, 1B queue, 24K table transactions; ~400,456,000 publishing executions, 1M API executions, 20 ms average response; total cost ~59 NOK per month. Domain logic uses discriminated unions and a 48‑hour usage‑rights buffer.

AI-written summary · about

" chapters (8)

1. 0:00Intro
2. 1:36Why Total Rewrite
3. 3:38Technical Stack
4. 4:09Workflow
5. 6:32Performance
6. 7:41Cost
7. 8:15Domain Specific Rule
8. 10:06Summary

" related talks

• 53mUsing F# on Azure Functions in ProductionNikolai Andersen · Build Stuff 2017
• 58mUsing F# on Azure Functions in ProductionNikolai Andersen · NDC Oslo 2017
• 01h 02mScaling Serverless F# with Azure FunctionsMathias Brandewinder · NDC Oslo 2017
• 01h 01mServerless Azure Data handlingBenjamin Kettner, Frank Geisler · Global Azure 2021
• 59mServerless Compute with Azure FunctionsMagnus Mårtensson · NDC Mini Copenhagen 2017
• 25mBuilding AI-Powered Applications with Azure Serverless Functions and C# .NETNaveen Chatlapalli · Commit Your Code 2025
• 47mAzure Functions PowerShell: The Missing ManualJustin Grote · PowerShell + DevOps Global Summit 2022
• 42mServer-less compute with .NET based Azure FunctionsAlex Thissen · .NET fwdays'18