Build a Large Language Model (From Scratch) & Systems Programming with Zig & Grokking Bayes.(COMBO)

Description

Build a Large Language Model (From Scratch)

Build a Large Language Model (From Scratch) is a practical and eminently-satisfying hands-on journey into the foundations of generative AI. Without relying on any existing LLM libraries, you’ll code a base model, evolve it into a text classifier, and ultimately create a chatbot that can follow your conversational instructions. And you’ll really understand it because you built it yourself!

Systems Programming with Zig

Systems Programming with Zig teaches you to write working, real-world code in Zig. It builds on your existing coding skills to introduce systems programming using the Zig language. You’ll explore everything from a low-level word ladder game to a CHIP-8 interpreter to a graphics engine with OpenGL. Each project is built from the ground up, with no libraries or imported frameworks! Chapter by chapter, you’ll learn how to deliver command-line utilities, networking with TCP and HTTP, interpreters, and graphics programming all through Zig’s unique lens.

Your expert guide to Zig is Garrison Hinson-Hasty, contributor to the Zig project and its wider ecosystem. Through fun anecdotes and silly scenarios, Garrison turns systems programming from “tedious” to “terrific.”

Grokking Bayes.

Grokking Bayes teaches Bayesianism through clear explanations, rich illustrations, and relatable examples. You’ll first build an intuition, and then translate that intuition into working code with Python tools like the PyMC library and ArviZ package. Along the way, you’ll explore how Bayesian ideas connect to modern AI, from uncertainty-aware deep learning to LLM applications.

Throughout, the book focuses on the skills for making better decisions: you’ll go from inference, to Bayesian decision theory, and even experimental design. Whether you’re a data scientist, AI practitioner, or curious learner, Grokking Bayes will give you the tools to make smarter decisions without needing to rely on long-run frequencies or reams of data without drowning you in probability theory and abstract math.