Towards AI (Medium)
AI Agents Spoke to One Another on a Public Message Board
5 mins
AnalysisDeep Reads.
Top-tier AI blogs, technical tutorials, and research analysis written by the people shaping the industry.
Last Brew Time: Sep 6, 2026, 10:33 AM PT
Featured Articles
Towards AI (Medium)
Your Model Memorized More Than You Think — and That’s One Bug, Not Four
7 mins
Research![ML@SCALE · 1:1 · Every training label was lying [Edition #4]](https://substackcdn.com/image/fetch/$s_!jGMl!,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2Fd2060521-ad96-49dd-bdb0-1668474b80bb_1408x768.jpeg)
Machinelearningatscale Substack
ML@SCALE · 1:1 · Every training label was lying [Edition #4]
7 mins
ResearchTowards AI (Medium)
What Do You Do With a Model That’s Too Big for Your GPU?
9 mins
TutorialTowards AI (Medium)
Codex Plugin for Claude Code: Choose, Install, and Run a Reliable Two-Agent Workflow
9 mins
Tutorial
Data4Sci Substack
LLMs Are the New Wikipedia
5 mins
Analysis
Kenhuangus Substack
Chapter 2: The KV Cache Frontier: Hybrid Compressed Sparse Attention (DeepSeek-V4 CSA/HCA) & MLA, Global Prefix Caching & KVShare
7 mins
ResearchTowards AI (Medium)
Multi-Agent Systems Are the New Microservices — and They’re Making the Same Mistakes
5 mins
Analysis
Developersdigest
GPT-6 In 7 Minutes
3 mins
News
Lenny's Newsletter
Why companies are becoming a series of loops | Anish Acharya (a16z)
5 mins
AnalysisTowards AI (Medium)
Muse Spark 1.3 Outperforms 5.6 Sol At Half The Price, How Did Meta Catch Up This Quick
3 mins
News
Thesequence Substack
The Sequence Radar - Issue 927: Last Week in AI: Model Madness: The Frontier Has a Refresh Button
3 mins
News
Highlearningrate Substack
Research Agents + Off-Peak Inference - 2026-09-06
3 mins
News
Arxiviq Substack
Normalized Low-Rank Adaptation
7 mins
ResearchOpenAI Engineering
Research acceleration: The view inside OpenAI
5 mins
Analysis
Developersdigest
Coding Agents Need Better Human Loops, Not Just Harder Benchmarks
5 mins
Analysis
Every
A Split Verdict on Fable vs. Astra
5 mins
AnalysisSimon Willison
Introducing GPT-6 Astra for developers
4 mins
News
Bretkerr Substack
Glassblowers of the AGI Age
5 mins
Analysis
Thezvi Substack
Claude Mythos 5.1 and Fable 5.1: Capabilities
5 mins
Analysis
Rickhigh Substack
Whoever Can Write the Memory File Edits the System Prompt: Harness Engineering - Prompt Injection Prevention IX
5 mins
Analysis