HeyJaredVanishing Gradients
A podcast for people who build with AI. Long-format conversations with people shaping the field about agents, evals, multimodal systems, data infrastructure, and the tools behind them.
- Indexed pieces, last 90 days
- 16
- Latest publication
- Sep 28, 2026
- Audience
- By platform
- Earliest in this view
- Jul 22, 2026
One publication, every platform
Open a platform to follow it on its own.
Why these are grouped
They share a verified publisher identifier or are linked from the publication’s official site. People stay separate from their publications.
Latest pieces
Agentic Data Science: Specify the Work, Verify the Claim (opens the original)
Read excerpt
When an AI agent can explore a dataset, choose a modeling approach, run the analysis, and explain its findings, what should the data scientist do?Traditionally, data scientists chose each step and implemented much of the analysis themselves. Agentic data science changes that division of work: we can delegate an investigation, including methodological choices, while shaping the question, supplying relevant expertise, and challenging the evidence it produces. For AI-native data scientists, choosin
Does Your AI Agent Actually Work? A Guide to Evals (opens the original)
Read excerpt
You have a collection of documents, so you build a retrieval system. Then you add an agent that can reason over them. The demo looks promising, and the team starts suggesting improvements: a different model, better chunking, more tools.But there’s a question you haven’t answered yet: how do you even test whether it works?Before choosing an evaluator, you need to know what the product is supposed to do, who will use it, and what a successful result looks like. Those decisions give you something c
The 6 Most Common AI Agent Mistakes (opens the original)
Read excerpt
How to choose where an agent belongs, diagnose what goes wrong, and test whether your changes help.You have a collection of documents, so you build a retrieval system. Then you add an agent because it can reason over the whole collection. A new model comes out, so you switch. Six months later, you’ve built a lot of software, but you still can’t answer a basic question: does it solve the problem you started with?It’s easy to keep adding capabilities when we don’t know what’s failing. An agent mig
How to Build a Coding Agent with Nico (Amp Code) (opens the original)
Read excerpt
“It’s an LLM, a loop, and enough tokens.”—Nicolay Gerold, Amp CodeSo why does a coding agent forget your instructions, keep reading tiny pieces of a file, or struggle with an edit another model handles easily?Nicolay Gerold (Amp) joins Hugo to take that loop apart and explain the harness around it: the software that executes tools, manages context, and lets you steer the agent’s work. Nico builds this machinery for Amp, a coding agent that works across software projects.“Every compo
Beyond Navier–Stokes: Who Controls Scientific Discovery? (opens the original)
Read excerpt
Is the current furore in mathematics the canary in the coalmine for experimental science and knowledge work?Science without understanding?I recently went back to Dresden for the 25th birthday of the Max Planck Institute (MPI) for Cell Biology and Genetics, where I did part of my postdoc. The MPI was founded to research the physical and biological mechanisms of cells to bridge the gap between the molecular and tissue scales. At the anniversary conference, Michael Bronstein (DeepMind Professor of
Publishing over time
Last 90 days. Choose a month to open its work.
Recurring subjects
Named in the text we hold. One piece can cover several.
Audience
Each platform is measured on its own. Audiences can overlap.
Newsletter
No verified audience measurement yet.
Podcast
No verified audience measurement yet.
About this data
Counts cover the work we have indexed. Tone needs enough text and a confident classification. Excerpts and episode notes are not full articles or transcripts.
Identity or attribution wrong? Suggest a correction.