Linear Digressions

Linear Digressions

Katie Malone
Country USA
Genres Technology
Language EN
Episodes 312
Latest 16.11.2026

Linear Digressions demystifies artificial intelligence and machine learning for the intelligently curious. Host Katie Malone explores complex topics in AI, making them accessible and engaging. The podcast covers a wide range of subjects from algorithms to real-world applications.

Episodes

  • How to Stay Relevant as a Data Scientist in the Age of AI | Interview with Dr. Jon Krohn
    How to Stay Relevant as a Data Scientist in the Age of AI | Interview with Dr. Jon Krohn 16.11.2026 43m
    Staying relevant as a data scientist when AI is reshaping the field isn't just a career question — it's an existential one. Dr. Jon Krohn, neuroscientist-turned-data-science-educator, podcaster, and consultant, sits down to share how he thinks about navigating that challenge. From his unconventional path out of academia to the advice he gives practitioners trying to future-proof their careers, this conversation is packed with hard-won perspective. If you've ever wondered what it takes to keep growing in a field that won't stop moving, this one's for you. Links: Jon Krohn's website https://www....
  • GPU Architecture for LLM Inference: Memory, Bandwidth and Hardware Explained
    GPU Architecture for LLM Inference: Memory, Bandwidth and Hardware Explained 09.11.2026 19m
    GPUs are the engines powering every AI conversation you've ever had — but what's actually happening inside those chips when an LLM generates a response? Understanding GPU fundamentals turns out to be surprisingly useful for making sense of why inference works the way it does, from the prefill stage that processes your entire prompt at once to the serial decode step that produces tokens one by one. If you want to understand the real constraints shaping modern AI systems, this is where to start.
  • KV Cache Fundamentals
    KV Cache Fundamentals 02.11.2026 24m
    The unsung hero of every LLM conversation you've ever had? A clever optimization called the KV cache. Self-attention is computationally brutal — every token attending to every other token means costs that scale as N², and that blowup gets worse with every turn of a conversation. The KV cache is what keeps that from becoming completely intractable at scale. Today's episode goes deep on how it works, why inference is a harder and more interesting problem than it gets credit for, and what's really happening under the hood when you chat with an LLM.
  • AI Inference 101
    AI Inference 101 26.10.2026 27m
    Everyone talks about training the model — the data, the compute, the alignment magic. But what happens after the weights are frozen and millions of users start hammering it with prompts? That's inference, and it's a surprisingly deep rabbit hole of engineering challenges that rarely gets the spotlight it deserves. This week I kick off a new series unpacking the full inference ecosystem — from the moment a model completes training to the moment tokens land in your hands.
  • Back Next Week
    Back Next Week 19.10.2026
    Back Next Week by Katie Malone
  • Back in Two Weeks
    Back in Two Weeks 12.10.2026
    Back in Two Weeks by Katie Malone
  • Mechanistic Interpretability: How Researchers Try To See Inside Models
    Mechanistic Interpretability: How Researchers Try To See Inside Models 05.10.2026 29m
    As AI systems grow more powerful, they grow harder to understand — and that tradeoff has never mattered more. Most interpretability tricks give us approximations from the outside, but mechanistic interpretability takes a different approach: cracking open the model itself to trace *why* it behaves the way it does. This episode digs into what makes mechanistic interpretability fundamentally different from other explainability methods, and why it might be the most promising path toward actually understanding what's happening inside modern LLMs.
  • LLM As A Judge
    LLM As A Judge 28.09.2026 30m
    Labeled data is expensive, slow, and painfully hard to come by — and when you're trying to evaluate whether an AI workflow is actually doing what you want, the problem gets even messier. How do you judge whether a chatbot response was helpful, honest, or hallucination-free at scale? This episode digs into using LLMs as judges: letting the models themselves evaluate the quality of AI outputs, from simple chat exchanges all the way to complex multi-step agent behavior.
  • The Impact of AI on Podcasting (Harvard Data Science Review Cross-Post)
    The Impact of AI on Podcasting (Harvard Data Science Review Cross-Post) 21.09.2026 29m
    Originally aired on the Harvard Data Science Review podcast. What can podcasting teach us about AI — and what can AI teach us about the future of podcasting? Katie Malone (that's our host) joins Jon Krohn of SuperDataScience for a conversation with Harvard Data Science Review editor-in-chief Xiao-Li Meng about both questions at once. They dig into what it means to cover a field that's moving this fast, who these shows are really for, and why that sweet spot between "too high level" and "too in the weeds" is so hard — and so worth chasing.
  • Better Know A Benchmark: ExploitGym
    Better Know A Benchmark: ExploitGym 14.09.2026 32m
    When OpenAI's frontier models were caught hacking Hugging Face's servers, most people assumed they were hunting for answer keys. The real story is stranger and more unsettling. Katie and Phoebe unpack ExploitGym — the cybersecurity benchmark at the center of the incident — and why agents are scored not just on whether they capture the flag, but on whether they used the specified vulnerability to get there. That nuance turned out to be load-bearing: the agents reverse-engineered the flags within the first hour, then spent days attacking Hugging Face to learn how the LLM judge worked so they cou...
  • Constitutional AI
    Constitutional AI 07.09.2026 31m
    How do you teach a model the difference between helpful and harmful when it has no inherent sense of either? This episode dives into Constitutional AI, Anthropic's framework for training AI systems to be both useful and safe by giving them an explicit set of principles to reason from. It's a fascinating look at how alignment research is evolving beyond simple human feedback — and what it means to give an AI something like a conscience. Links: Anthropic, "Constitutional AI: Harmlessness from AI Feedback" (2022) https://arxiv.org/abs/2212.08073 Claude's Constitution https://www.anthropic.com/con...
  • A Data-Driven Reality Check on AI in Business (Interview with Tom Davenport, Babson College)
    A Data-Driven Reality Check on AI in Business (Interview with Tom Davenport, Babson College) 31.08.2026 40m
    Tom Davenport — the man who called data science "the sexiest job of the 21st century" — is back with a reality check on AI. As one of the most seasoned observers of how businesses actually adopt transformative technology, Davenport brings a rare, well-calibrated perspective to the AI hype cycle. Is this moment genuinely different from past paradigm shifts, or are we pattern-matching to a familiar story? Katie sits down with her old colleague to find out what's really happening when companies try to put AI to work.
  • Understanding AI Text Watermarking
    Understanding AI Text Watermarking 24.08.2026 29m
    Anthropic just announced they're baking invisible watermarks directly into Claude's generated text — and while everyone else was busy having opinions about it, we were busy asking the more interesting question: how does it actually work? Turns out it's not hidden Unicode characters or first-letter secret codes — it's something far more elegant, operating at the level of word choice itself. We dig into Google DeepMind's SynthID text approach, published in *Nature* in 2024, to understand the clever statistical machinery behind watermarking language model outputs without anyone being the wiser.
  • Better Know a Benchmark: Humanity's Last Exam
    Better Know a Benchmark: Humanity's Last Exam 17.08.2026 23m
    Humanity's Last Exam was designed with a bold premise: questions that human experts can answer, but AI models can't. Originally dubbed "Humanity's Last Stand," this benchmark is a massive academic collaboration — hundreds of contributors, thousands of fiendishly hard questions spanning a wild range of domains. In this Better Know a Benchmark installment, we unpack what HLE is actually testing, how it was built, and what it means when a model finally starts cracking it.
  • A Scientific Deep Dive into Overconfident LLMs: Interview with Kaitlyn Zhou (Cornell)
    A Scientific Deep Dive into Overconfident LLMs: Interview with Kaitlyn Zhou (Cornell) 10.08.2026 33m
    When a language model tells you it's absolutely certain, is it actually more likely to be right? Kaitlyn Zhou's research says: not necessarily — sometimes confident phrasing correlates with *worse* accuracy, echoing a very human Dunning-Kruger effect. In this conversation, Kaitlyn (soon an assistant professor at Cornell) walks through why LLMs talk this way in the first place — tracing the tendency back through training data and the RLHF annotation process, where it turns out humans don't love confidence so much as they punish uncertainty — and what that does to the person on the other end of...
  • Reasoning Models: When LLMs Went Beyond Fancy Autocomplete
    Reasoning Models: When LLMs Went Beyond Fancy Autocomplete 03.08.2026 25m
    Reasoning models don't just answer your question — they *think out loud* first. In this episode we dig into the class of AI models that generate intermediate chains of thought before arriving at a final answer, exploring how the internal reasoning process works. Are these models genuinely "thinking," or is something else going on under the hood?
  • Distillation, or, How to Steal a Model
    Distillation, or, How to Steal a Model 27.07.2026 23m
    This week we’re covering model distillation: the technique of using a large "teacher" model's outputs to train a smaller, cheaper "student" model that mimics it. They cover the two big reasons labs do this — making lighter, faster, more focused models for specific tasks, and the more contentious use case of effectively copying a rival's flagship model by hammering its API with questions (with a callback to the old Bing/Google search controversy). They also get into why it's so hard to prove distillation happened, why some models occasionally introduce themselves as "Claude," and a surprisingly...
  • Invisible LLM Failures and AI Fluency with Chris Potts (Stanford)
    Invisible LLM Failures and AI Fluency with Chris Potts (Stanford) 20.07.2026 41m
    What happens when a Stanford linguistics professor turns his attention to AI chatbots — and the surprisingly invisible ways humans misunderstand them? Chris Potts joins the show to unpack the hidden failure modes in how we interact with AI, what it really means to become a more fluent user, and why these language-wielding systems are genuinely alien in ways we're only beginning to reckon with. His perspective sits at a rare intersection of linguistics, cognition, and machine learning — and it shows.
  • Still summer break: back next week
    Still summer break: back next week 13.07.2026
    Still summer break: back next week by Katie Malone
  • Summer break: back soon
    Summer break: back soon 06.07.2026
    Summer break: back soon by Katie Malone

Popular in

The podcast also appears in the podcast charts of these countries.