Ahead of AI logo

I'm Sebastian: a machine learning & AI researcher, programmer, and author.

Sebastian Raschka Profile Picture
As Staff Research Engineer at Lightning AI, I focus on the intersection of AI research, software development, and large language models (LLMs).
Lightning AI logo
I used to hold a position as an Assistant Professor of Statistics at the University of Wisconsin-Madison (on a tenure track from 2018-2025). However, with a heavy heart, I recently resigned in 2023 to concentrate fully on my work at the Lightning AI startup, which I had joined in January 2022.
UW-Madison logo
Moreover, I love open-source software and am a passionate contributor. Next to coding, I also love writing and covering the latest deep learning & AI research in Ahead of AI.
Ahead of AI Logo
I also have a passion for education and am the author of several books!

News



Nov 21, 2024

I recently had the pleasure of joining Neil to discuss a wide range of topics on AI Stories. It was a fun chat, and our conversation spanned lots of (hopefully interesting) topics:

  • Life in both academia and industry
  • LLM training and cloud computing
  • The ins and outs of implementing and building LLMs from scratch
  • Transitioning from GPT architectures to Llama 3.2
  • Long-context LLMs vs Retrieval-Augmented Generation (RAG)
  • How multimodal LLMs work

AI Stories Podcast



Sep 16, 2024

After 1.5 years of hard work, "Build A Large Language Model (From Scratch)" is finally published! Print and ebook copies are available on Manning's website. And the book is also available on Amazon.

Build an LLM from Scratch release



July 31, 2024

II'm excited for my first PyTorch conference! Can't wait to meet the community and chat about the latest in AI and LLM developments!

PyTorch Keynote Talk



May 17, 2024

I'll be giving a 1-hour ACM Tech Talk on June on "Understanding the LLM Development Cycle: Building, Training, and Finetuning". This talk will guide you through the key stages of developing large language models (LLMs), from initial coding to deployment. This talk is virtual, and you can register for free here.

ACM Tech Talk on Developing LLMs



May 15, 2024

Last week, I sat down with Hugo Bowne-Anderson from the Vanishing Gradient to record a podcast all about LLMs. We ended up covering the entire LLM lifecycle, what type of skills you need to work with them, what type of resources and hardware, prompt engineering vs finetuning vs RAG, and more! (Plus, I gave a 30 min live demo finetuning an LLM for classification). You can find a link to the podcast here and a YouTube video version here.

LLMs Vanishing Gradient Podcast



Mar 25, 2024

I somehow made it to GitHub's top-1 spot on the Trending Developers list. I've been coding on GitHub quite consistently for about 12 years, but I honestly never expected to find myself up there! As a passionate coder and open-source developer, this is perhaps the nicest compliment for me so far!

GitHub Trending



Mar 19, 2024

I had the pleasure of sitting down with Jon Krohns on the SuperDataScience podcast for what turned into a roughly 2-hour mega interview on LLMs. We covered pretty much everything!
LLM Mega Interview

  • Developing efficient open-source codebases for finetuning and pretraining LLMs
  • A developer-friendly codebase for experimenting with LLM research ideas
  • Accelerating PyTorch code with Fabric
  • Scaling research: Running multi-GPU and multi-node AI experiments with Lightning Studios
  • LLM architecture deep dives: OLMo and Gemma LLMs
  • From 70B to 3B; and generalized vs specialized LLMs
  • LoRA vs DORA: Efficient LLM finetuning
  • Writing books about LLMs


Feb 8, 2024

I'll be giving a 3.5 hour deep learning workshop at PyCon 2024 in May. This tutorial is aimed at Python programmers new to PyTorch and deep learning. However, even more experienced deep learning practitioners and PyTorch users may be exposed to new concepts and ideas when exploring other open source libraries to extend PyTorch.
It's my first PyCon, and I'm very excited!

PyTorch at PyCon 2024




All News: 20232022202120202019