Sebastian Raschka
Person on 4 sites
Sebastian Raschka is an LLM research engineer and the author of books on machine learning, including Build a Large Language Model (From Scratch).
-
4 facts on wiqqi (opens wiqqi.app)
- Now: LLM research engineer and author (as of November 2026) (opens wiqqi.app)
- Before: Professor, Department of Statistics, University of Wisconsin-Madison; senior engineer, Lightning AI (opens wiqqi.app)
- Book: Build a Large Language Model (From Scratch) (opens wiqqi.app)
-
4 channels on ppll (opens ppll.app)
-
GitHub: github.com/rasbt (opens github.com)
“reasoning-from-scratch — Implement a reasoning LLM in PyTorch from scratch, step by step”
-
x.com: @rasbt (opens x.com)
“The complete reasoning-from-scratch video series is now up (and linked) here: https://t.co/5IOHLTCIGt (Taking a short break but will be back with LLM model and research updates in November!)”
-
Site: sebastianraschka.com (opens sebastianraschka.com)
“Building Reasoning Models from the Ground Up: A Full Video Course”
-
GitHub: github.com/rasbt (opens github.com)
-
16 beliefs on korrents (opens korrents.com)
-
Reasoning models are an upgrade from conventional LLMs achieved by adding post-training via reinforcement learning and optionally inference scaling. (opens korrents.com)
“A reasoning model is essentially an upgrade from a conventional LLM, that is, adding more post-training via reinforcement learning with verifiable rewards (and optionally inference scaling techniques) to make it more…”
-
Frontier language models will remain reasoning models trained with RLVR-style chain-of-thought for the foreseeable future. (opens korrents.com)
“GPT-6 Astra (and likely any LLM in the foreseeable future) is still a reasoning model.”
-
Computer-use capability will be a major focus of LLM and coding-agent development in the coming months and years. (opens korrents.com)
“I expect the upcoming months (or years) also to be an era of computer use refinement on both the LLM and the agent harness layer.”
-
Reasoning models are an upgrade from conventional LLMs achieved by adding post-training via reinforcement learning and optionally inference scaling. (opens korrents.com)
-
7 recommendations on rcmnd (opens rcmnd.app)
-
Codex (opens rcmnd.app)
“I have to admit that I still primarily alternate between Codex and Claude Code as my daily drivers, for now”
-
Claude Code (opens rcmnd.app)
“I have to admit that I still primarily alternate between Codex and Claude Code as my daily drivers, for now”
-
Ollama (opens rcmnd.app)
“I am not affiliated with any of the tools mentioned in this article, but one nice thing about Ollama is that they also optionally support open-weight models hosted in the cloud”
-
Codex (opens rcmnd.app)