Underfitted

Home
Notes
Archive
Leaderboard
About
How to evaluate a RAG application automatically
You'll learn how to build and evaluate a simple RAG application that answers questions from a website.
2
Building a RAG application from scratch using LangChain and the OpenAI API
You'll learn how to build a simple application to answer questions from YouTube videos using an LLM.
1

February 2024

Be careful with state-of-the-art models
The three stages of model building and the consequences of playing stupid popularity contests.

January 2024

Test-Time Augmentation
Using data augmentation at inference time to make better predictions.
Adversarial validation
How to determine whether your training and test data come from the same distribution.
7
My trick for framing machine learning problems
Inversion, the haystack principle, and how they'll help you win every project.
2
Impossible to fail
You don't win with better algorithms or data. You win by framing problems in a way that makes it impossible to screw things up.
2

July 2023

Building software with a LLM is not what they say
Five problems you'll have to deal with when using Large Language Models.
4

May 2023

We integrated ChatGPT with the Spot Robot
A quick overview about why and how we built the integration and what will happen next.

April 2023

I Met An Old, Unbelievable Person
Most people know nothing about probabilities and fall victims of cherry picked stories every day.
4
Five Reasons Production Machine Learning is Different From Everything You’ve Seen
Unfortunately, traditional education focuses on a subset of what's required to build a Machine Learning system.

March 2023

Stop Bragging About How Many Books You Read
Nothing can compete with getting your hands dirty and doing the work. If you want to learn, this is the only way.
© 2024 Santiago L. Valdarrama
Privacy ∙ Terms ∙ Collection notice
Start WritingGet the app
Substack is the home for great culture