My Life as a Tech Transfer Monad
summary
Argues for making uncertainty a first-class language feature and shows concrete work: a probability monad implemented in PHP/Hack, sampling via multishot coroutines, and Monte Carlo execution that runs programs millions of times to build distributions. Demonstrates linear regression learned from four training points that peaks at 8 for input 4, and a WebPPL-style example compiled to run on Facebook’s VM. Describes teams building runtime, type/debug support, and ML-driven developer tools (autocomplete, fix suggestion), and reports those components are implemented and prototype‑working.
AI-written summary · about
chapters (23)
- 0:00Intro
- 2:05Title
- 3:16Monnet
- 5:40The Curse of Computer Science
- 8:05The Dream World
- 11:05Relational Algebra
- 12:43Sequel
- 13:25Relational Databases
- 14:55Programming Language Landscape
- 16:10Distractions
- 17:43Learning Based Development
- 20:40Old Math
- 22:43Automated TDD
- 24:25Backpropagation
- 26:45Metric Spaces
- 28:09Linear Regression
- 32:51Code
- 35:43Probability Distribution Sampling
- 41:23Embracing uncertainty
- 44:23Reinforcement learning
- 47:03Program of the future
- 48:58Example
- 53:26Machine Learning for Developers
related talks
59mDo or don’t There’s no try Or is there The power of monads explained Sort of
42mMonads you already use without knowing it
30mMonad for dummies
43mM-M-M-Monads!
01h 04mHaskell - A Peek Inside the Ivory Tower
41mLiving the Monadstic Life
26mUP YOUR MTL GAME WITH CLASSY OPTICS
44mIO monad & Error management: From Exceptions to Cats MTL