In this episode I briefly explain the concept behind activation functions in deep learning. One of the most widely used activation function is the rectified linear unit (ReLU). While there are several flavors of ReLU in the literature, in this episode I speak about a very interesting approach that keeps computational complexity low while improving performanc ... Show More
Oct 26
I wrote a chess engine and still can't play chess: AI coding and skill decay (Ep. 317)
I wrote a chess engine that beats real players. I still can't play chess. Developers who hand their coding to AI are heading the same way: they can build the machine but can't play the game. We go through the evidence (Anthropic, METR, Microsoft/CMU), the best counterarguments, a ... Show More
34m 48s
Oct 20
"Rust Is Amazing, as Long as Nobody Looks at It" - Fact-Checking DHH (Ep. 316)
DHH told Rails World that hand-written code is over, that HEY's backend is being AI-rewritten in Rust ("amazing, as long as you never look at it"), and that developer happiness no longer matters. This episode checks his statements. Struggling with AI and Software? Here is my new ... Show More
22m 56s
Oct 13
AI coding agents have a Mythical Man-Month problem (Ep. 315)
Fifty years ago Fred Brooks showed why adding more people to a late project makes it later. Today we do the same thing with tokens — more context, more autonomy, same myth, different currency. Struggling with AI and Software? Here is my new book available for purchase on https:// ... Show More
17m 29s
Nov 2022
The practicalities of releasing models
Recently Chris and Daniel briefly discussed the Open RAIL-M licensing and model releases on Hugging Face. In this episode, Daniel follows up on this topic based on some recent practical experience. Also included is a discussion about graph neural networks, message passing, and tw ... Show More
37m 19s
May 2018
Practical Deep Learning with Rachel Thomas - TWiML Talk #138
In this episode, i'm joined by Rachel Thomas, founder and researcher at Fast AI. If you’re not familiar with Fast AI, the company offers a series of courses including Practical Deep Learning for Coders, Cutting Edge Deep Learning for Coders and Rachel’s Computational Linear Algeb ... Show More
44m 19s
Mar 2025
180: Reinforcement Learning
Patrick and Jason introduce reinforcement learning and place it alongside supervised and unsupervised learning. They cover Q-learning, SARSA, policy gradients, actor-critic methods, PPO, imitation learning, and why training and evaluating RL systems is so challenging.
1h 52m
May 2025
Hugging Face Explained: The AI Toolkit Everyone’s Talking About
In this episode we cover Hugging Face—what it is, why it matters, and how it's powering some of the biggest AI models today. We’ll break down their open-source tools, transformers library, and how developers and companies are using them to build the future of AI.“AI Hustle” PODCA ... Show More
12m 11s
