The Deep End latest|26|25|24|23|22|21|20|19|18|17|16
09.24 [2609.24972] RRSI: Regularized Recursive Self-Improvement of Agent Harnesses (arxiv.org)
09.23 tokens too cheap to meter (jyn.dev)
Claude discovers a novel enzyme system \ Anthropic (www.anthropic.com)
Introducing Strands harness: frontier performance with 28% lower token cost... (strandsagents.com)
Meet Stripe's Knowledge AI Platform | Stripe Dot Dev Blog (stripe.dev)
09.22 Open-Weight Inference Economics & GPU Useful Life | Ornn Data (data.ornn.com)
Step 2. Token Embedding (poloclub.github.io)
09.20 Trying the Software factory pattern. | Irrational Exuberance (lethain.com)
阶跃星辰 (www.stepfun.com)
the senior engineer death spiral • Solving the decision problem (sunilpai.dev)
Pirate Face - Turn AI into torrents that live forever (pirateface.co)
Poster Prompts | ‘ERE I AM - JH! (ukslim.github.io)
09.19 I Built Non-Autoregressive Decision Models with RL a Year Ago | Hacker News (news.ycombinator.com)
Laya — 33ms Multilingual System 1 Decision Engine with Calibrated Probabilities (laya.convaiinnovations.com)
09.18 AI chatbots are becoming experts at changing people's minds. What's their... (www.science.org)
Needle 2 - The 14 MB Agentic LLM for Tiny Devices | Cactus (cactuscompute.com)
[2609.20804] An Empirical Study of Harness Design for Coding Agents (arxiv.org)
09.17 The Anthropic Glasswing Receipts Are Starting to Trickle In | Blog | VulnCheck (www.vulncheck.com)
Meat proxy - sf-isms (sfisms.org)
Towards Self-Driving Codebases | Detail (blog.detail.dev)
Economic Policy for AGI — DeepMind Institute (institute.deepmind.com)
Toward Recursive Self-Improvement: How GLM Built Its Own Inference Infrastructure (z.ai)
[2609.18842] Infinite-Parameter LLMs: Generating and Adapting Weights from Live Data (arxiv.org)
LLM Classification Is Feature Engineering | Minimally Sufficient (minimallysufficient.com)
For AI leaders Doom is a form of hype! - Erkan's Field Diary (erkansaka.net)
09.16 [2511.07885] Intelligence per Watt: Measuring Intelligence Efficiency of Local AI (arxiv.org)
Apple Reference Image: A New Approach for Verified Photography - Apple... (security.apple.com)
09.15 Introducing System One Models and Jev - TypeSafe AI Blog (typesafe.ai)
25 Years of Mass Surveillance Is Enough - Schneier on Security (www.schneier.com)
Inside the Inference Hardware Revolution Of 2026 - IEEE Spectrum (spectrum.ieee.org)
https://orib.dev/gefs.pdf (orib.dev)
09.14 [2609.12303] Breaking the Token Ceiling: Distilling Smaller, Stronger Byte Models (arxiv.org)
[2609.11873] The Last AI Built by Humans: Toward Genuine Recursive Self-Improvement (arxiv.org)
The Last AI Built by Humans: Toward Genuine Recursive Self-Improvement |... (academy.dair.ai)
Distributed Systems Classics (nvartolomei.com)
GitHub - rush86999/atom: Atom Agent, Open-Source Governed AI Agent Platform... (github.com/rush86999)
Why we built Pion | Andon Labs (andonlabs.com)
Foundation Model Engineering (sungeuns.github.io)
Claude Fable 5.1 Solves the Cyphral Distich (www.vals.ai)
09.13 Countering misuse of AI: September 2026 / Anthropic \ Anthropic (www.anthropic.com)
Getting 50 GB/s Back Out of the ANE | Eileen Yoon (eiln.github.io)
Measuring Reward-Seeking by Instilling Contrastive Beliefs (alignment.openai.com)
[2607.18966] Measuring Reward-Seeking via Contrastive Belief Updates (arxiv.org)
Houthis Used Claude Code to Develop Missile Guidance Software: Anthropic ·... (clashreport.com)
09.12 Real-SWE Benchmark — Specific Labs (withspecific.com)
AI researchers debate how close we are to recursive self-improvement (www.dwarkesh.com)
09.11 So you want to use OpenRouter? — Mo Moustafa (mmoustafa.com)
Declaration — Math and AI (mathandai.org)
Global Glacier Extinction Explorer (glacierextinction.com)
09.10 To write non-fiction, draw the trunk, then the rest of the tree (devz.cl)
>10x More Efficient Pretraining — Magic (magic.dev)
What do Visa and Mastercard do? An intro to card networks @ tautology.town (tautology.town)
The Future of Humanoid Robots (www.berginsight.com)
Scenarios for our Economic Future \ Anthropic (www.anthropic.com)
From Models to Machines: The Physical AI Market Map (makecents.substack.com)
09.08 On the Navier–Stokes Millennium Prize Problem | OpenAI (openai.com)
[2504.02246] C*: Unifying Programming and Verification in C (arxiv.org)
09.07 GitHub - timgordontg/engrim: The Universal Cross-Model Episodic Memory... (github.com/timgordontg)
Ask HN: How do you manage skills files? | Hacker News (news.ycombinator.com)
09.06 IBM Quantum Nighthawk r2—more circuits, faster | IBM Quantum Computing Blog (www.ibm.com)
An Alien Mind | OpenAI (openai.com)
Research acceleration: The view inside OpenAI | OpenAI (openai.com)
[2609.03344] Large-Language Models as a Cognitive Virus (arxiv.org)
AI, tools and transformation — Benedict Evans (www.ben-evans.com)
Cloud in a Bottle: making self-hosting accessible to everyone | Cloud in a Bottle (cloudinabottle.org)
09.05 AI Milestones: a sourced timeline of achievements in artificial intelligence (achievements.ai)
PC Build Guide: Compare Parts by Budget | Logical Increments (www.logicalincrements.com)
LAN Party House (lanparty.house)
Atisbo — One ordered list of what to do next (www.atisbo.dev)
Portal by Spotify cut my Claude Code token usage by 90% | Spotify Engineering (engineering.atspotify.com)
GPT-6 Astra review: code review gains, privacy, and cost (www.coderabbit.ai)
AI handles incidents, engineers lose touch with their systems — Sylvain Kalache (www.sylvainkalache.com)
Announcing Artificial Analysis Intelligence Index v4.2 | Artificial Analysis (artificialanalysis.ai)
FLT: Anthropic has beaten me to it | Xena (xenaproject.wordpress.com)
Corporate America is getting hooked on open-source AI | Hacker News (news.ycombinator.com)
09.04 Corporate America Is Getting Hooked on Open-Source A.I. - The New York Times (archive.is)
The Two Abstractions of System Design: Hide or Reduce (muratbuffalo.blogspot.com)
Programming Z3 (z3prover.github.io)
Formalizing Fermat's Last Theorem \ Anthropic (www.anthropic.com)
Discovery of a new OpenAI agent message board (collusion.wiki)
Introducing K2 Horizon: Frontier Performance, Radically Open (ifm.ai)
08.31 tmpout_v (tmpout.sh)
08.30 METR and Redwood Offer Holy #%^@ Postmortem Of The HuggingFace Hack | Don't... (thezvi.wordpress.com)
The Hugging Face attack surprised me - by Ajeya Cotra (www.planned-obsolescence.org)
NFC Energy-Harvesting PCB Business Card with an MCU | Wilson Harper (wilsonharper.net)
Creepy crawlies — Konstantin Ryabitsev (people.kernel.org)
[1804.07389] Longest Straight Line Paths on Water or Land on the Earth (arxiv.org)
08.29 blueferry/PROTOCOL.md at main · erikwb/blueferry (github.com/erikwb)
Lakr233/vphone-cli (github.com/Lakr233)
08.28 Brief independent investigation of agents’ behavior, reasoning and... (metr.org)
Meet Rosalind Workbench: Empowering every scientist to be their own research... (developers.openai.com)
When Memory Becomes Production State | Jason Doyle (jasondoyle.ie)
LLMs Don't Replace Classical ML — They Feed It | Bilanc (bilanc.co)
The Analytical AI Handbook - Sutro Handbook (handbook.sutro.sh)
Stronger with every update: How we’re making Chrome and the web safer in the AI Era (blog.google)
M-Trends 2026: Data, Insights, and Strategies From the Frontlines | Google Cloud Blog (cloud.google.com)
Introducing Index: Building The World’s Largest and Most Diverse Physical Dataset (www.figure.ai)
08.27 SenteLabsAI/OpenExecutive: AI-powered virtual executive team — a single... (github.com/SenteLabsAI)
Nvidia Has Been in Talks to Buy Hugging Face for More Than $13 Billion -... (www.businessinsider.com)
08.26 The Hugging Face incident and the road ahead | OpenAI (openai.com)
Qwen (qwen.ai)
GLM-5.3-Flash: Frontier Intelligence, Flash Cost (z.ai)
08.24 What is a Harness? (earendil.com)
How Complex Systems Fail (1998) | Hacker News (news.ycombinator.com)
How Complex Systems Fail (how.complexsystems.fail)
What Is a Harness? | Hacker News (news.ycombinator.com)
How I Find Problems to Solve as a Staff Engineer - Lalit Maganti (lalitm.com)
My agent.md to improve LLM-assisted code quality (fabiensanglard.net)
Anthropic’s best AI model struggles to attract users as cheaper tools thrive (archive.ph)
08.23 JIT Compiling Code in 5μs - malisper.me (malisper.me)
08.22 A Friendly Introduction to Racket – geometridae (geometridae.bearblog.dev)
08.20 [2608.12645] Jagged Judges: Epistemic Stability Under Silence, Pressure, and... (arxiv.org)
08.19 Being ambitious and being a dad | Nicholas Charriere (nicholascharriere.com)
08.18 The Amazon tax | Seth's Blog (seths.blog)
08.17 Qwen 3.8 27B is excellent, but it defaults to wildly overthinking things (simonwillison.net)
08.16 Manfred Bergmann | Software Development | Blog (nnamgreb.de)
08.15 [2608.11095] Why Does CLAUDE.md Keep Growing? Catastrophic Remembering in... (arxiv.org)
08.14 GitHub - lajosdeme/mole: A deep-research agent with an enforced budget,... (github.com/lajosdeme)
08.13 Ordinary Abundance (ordinaryabundance.com)
GitHub - xoreaxeaxeax/skitter-creek-bath-salts: Unlocking _everything_ on... (github.com/xoreaxeaxeax)
DeepSeek Harness developer preview: Everything is a plugin (deepseek.com)
[2607.09424] A Sovereign, Open-Source Foundation Model for German and English (arxiv.org)
08.11 Stowaway — Ride along with the planes and satellites overhead (stowaway.live)
Why Go is an Ideal Language for AI-Assisted Software Engineering - Google... (developers.googleblog.com)
08.10 Code Mode: the better way to use MCP | Cloudflare Blog (blog.cloudflare.com)
[2607.29380] The Tragedy of the Cognitive Commons: How AI Could Disrupt the... (arxiv.org)
Ask HN: What are you working on? (August 2026) | Hacker News (news.ycombinator.com)
The Philippines' big offshoring industry is growing despite AI | Hacker News (news.ycombinator.com)
08.09 .:: The Hacker's Renaissance - A Manifesto Reborn ::. (phrack.org)
Is it all just vapourware? (kirahowe.com)
An agentic development environment. (openchamber.dev)
There Are Magic Hexagons of Every Order (gukov.dev)
Making difficulty curves in games - Stuff Made By Dave (www.davetech.co.uk)
Improving Heuristics (www.redblobgames.com)
08.08 Managing AI Coding Costs at Scale | Databricks Blog (www.databricks.com)
08.07 Inside vLLM: Anatomy of a High-Throughput LLM Inference System - Aleksa Gordić (www.aleksagordic.com)
AMD acquires AI chip startup Taalas to boost inference performance by... (www.theregister.com)
08.06 SILO: S3-Compatible Object Storage by Pigsty (silo.pgsty.com)
Pareto front - Wikipedia (en.wikipedia.org)
Mario meets Pareto (www.mayerowitz.io)
The Entropy of A Markov Chain - by CasualPhysicsEnjoyer (chillphysicsenjoyer.substack.com)
[2510.01395] Sycophantic AI Decreases Prosocial Intentions and Promotes Dependence (arxiv.org)
08.05 Benchmarking Coding Agents on Databricks’ Multi-Million Line Codebase |... (www.databricks.com)
[2602.16763v3] When AI Benchmarks Plateau: A Systematic Study of Benchmark Saturation (arxiv.org)
Pi, Minimal and Performant | EARENDIL (earendil.com)
08.04 Harness Engineering for Self-Improvement | Lil'Log (lilianweng.github.io)
08.03 MirrorCode: What's the largest software project AI can complete on its own?... (epoch.ai)
08.02 Open Source Turtle Robot (OSTR) : 32 Steps (with Pictures) - Instructables (www.instructables.com)
Free Printable Paper Globes — Folding Paper Globes (foldingglobes.com)
16-bits.org - home of 8bitbubsy (16-bits.org)
Maze Toys — Online Mazes, 3D Corn Mazes & Puzzle Games (maze.toys)
08.01 Introducing Inkling-Small - Thinking Machines Lab (thinkingmachines.ai)
Gemini Robotics 2 brings whole body intelligence to robots — Google DeepMind (deepmind.google)
GitHub - yc-software/qm: Multiplayer agent harness for work · GitHub (github.com/yc-software)
GitHub - microsoft/flint-chart: 🪄 Flint is a visualization language that... (github.com/microsoft)
07.30 GitHub - funador/claude-code-merge-queue: The local merge queue for parallel... (github.com/funador)
07.29 aistack (aistack.imec-int.com)
[2607.25398] HANDBOOK.md: A Benchmark for Long-Context Agentic Instruction Following (arxiv.org)
07.28 Linear Attention, Visualized: From Mamba-2 to Kimi Delta Attention (KDA) |... (snowchord.com)
[2510.26692] Kimi Linear: An Expressive, Efficient Attention Architecture (arxiv.org)
Kimi K3 Architecture Notes | Sebastian Raschka, PhD (sebastianraschka.com)
You Could Have Come Up With Kimi Delta Attention | Doubleword (blog.doubleword.ai)
07.26 An Inside Look at the Relay Market Powering Token Resellers and Fraud | Vectoral (vectoral.com)
07.24 Boyd's Law, on steroids (www.linkedin.com)
07.23 Writing by Hand is Good for your Brain - Here's how to do it (nealstephenson.substack.com)
07.22 Hatchet · The startup's Postgres survival guide (hatchet.run)
Show HN: Bento - An entire PowerPoint in one HTML file... (news.ycombinator.com)
Security incident disclosure — July 2026 (huggingface.co)
07.21 [2304.13705] Learning Fine-Grained Bimanual Manipulation with Low-Cost Hardware (arxiv.org)
07.20 American AI is locked down and proprietary. It's losing. (werd.io)
Kimi K3, Qwen 3.8, and Anthropic's (potential) Unravelling (www.emergingtrajectories.com)
07.19 Hardware is not so hard - Chip Weinberger (chipweinberger.com)
07.18 DSLs Enable Reliable Use of LLMs (martinfowler.com)
07.17 Kimi K3 Tech Blog: Open Frontier Intelligence (www.kimi.com)
(WIP) A Little Bit of Reinforcement Learning from Human Feedback (rlhfbook.com)
07.16 Microsoft Comic Chat is now open source | Microsoft Open Source Blog (opensource.microsoft.com)
Command Line Interface Guidelines (clig.dev)
07.15 GitHub - CyrusNuevoDia/capn-hook: Dynamic memory for coding agents: chart... (github.com/cyrusNuevoDia)
https://spawn-queue.acm.org/doi/pdf/10.1145/3762989 (spawn-queue.acm.org)
arda tasci (eardatasci.github.io)
Inkling - Thinking Machines Lab (thinkingmachines.ai)
07.14 GitHub - Danau5tin/ai-trains-ai: RL-training an AI agent to RL-train AI... (github.com/Danau5tin)
Demis Hassabis has a plan to harness AI safely | Hacker News (news.ycombinator.com)
GitHub - hkjarral/AVA-AI-Voice-Agent-for-Asterisk: An open-source AI Voice... (github.com/hkjarral)
Guardian Angels: LLM Personalization for Productivity and Security · Gwern.net (gwern.net)
The zero-cost fallacy: Open source software in the agentic era | Thoughtworks (www.thoughtworks.com)
https://elasticity.institute/rsi-paper.pdf (elasticity.institute)
07.13 The 4-bitter Lesson | humans& (humansand.ai)
GitHub - strongdm/cxdb · GitHub (github.com/strongdm)
GitHub - strongdm/attractor: nlspec of StrongDM's Attractor, a... (github.com/strongdm)
How StrongDM’s AI team build serious software without even looking at the code (simonwillison.net)
07.12 Book: RISC-V System-on-Chip Design | Hacker News (news.ycombinator.com)
Digital Deli (www.atariarchives.org)
07.10 The Annotated JEPA | Elements of a Vector Space (elonlit.com)
07.07 The Annotated Transformer · 30 papers (30papers.com)
The Annotated Transformer (nlp.seas.harvard.edu)
07.06 A global workspace in language models \ Anthropic (www.anthropic.com)
[OpenWrt Wiki] OpenWrt One (openwrt.org)
Pulpie: Pareto-Optimal Models for Cleaning the Web | Feyn (usefeyn.com)
Aluminum foil ⁑ Dernocua (dernocua.github.io)
07.05 GitHub - langchain-ai/openwiki: OpenWiki is a CLI that writes and maintains... (github.com/langchain-ai)
Introduction to Compilers and Language Design | Prof. Douglas Thain at Notre Dame (dthain.github.io)
activegraph — A persistent world for long-running agents (activegraph.ai)
[2605.21997] The Log is the Agent: Event-Sourced Reactive Graphs for... (arxiv.org)
GitHub - smartcomputer-ai/lightspeed: Deterministic agent harness for... (github.com/smartcomputer-ai)
What ORMs have taught me: just learn SQL (wozniak.ca)
Fast Software, the Best Software — by Craig Mod (craigmod.com)
07.04 [2606.09118] ComplexConstraints and Beyond: Expert Rubrics for RLVR (arxiv.org)
jamesob/local-llm: Everything I know about running LLMs locally (github.com/jamesob)
Leanstral 1.5: Proof Abundance for All (mistral.ai)
07.02 Senior SWE-Bench (senior-swe-bench.snorkel.ai)
07.01 ZCode - Simple, Fast, Vibe‑Ready | Official Harness for GLM-5.2 (zcode.z.ai)
06.27 Fintech Engineering Handbook (w.pitula.me)
[2508.12631] Beyond GPT-5: Making LLMs Cheaper and Better via... (arxiv.org)
workweave/router: Model router for agentic systems. Routes every prompt to... (github.com/workweave)
Ultrasound imaging of the brain — Aleph (alephneuro.com)
[2305.18274] Reconstructing the Mind's Eye: fMRI-to-Image with Contrastive... (arxiv.org)
06.25 Prince Chazz (princechazz.com)
GitHub - inkeep/open-knowledge: Beautiful, AI-native markdown editor and LLM... (github.com/inkeep)
IBM Debuts World’s First Sub-1 Nanometer Chip Technology (newsroom.ibm.com)
The annotated PyTorch training loop — idlemachines (idlemachines.co.uk)
06.24 FUTO Swipe (swipe.futo.tech)
The Coming Loop | Armin Ronacher's Thoughts and Writings (lucumr.pocoo.org)
06.23 The truth about being a manager | sofiakodar.github.io (sofiakodar.github.io)
TikZ Editor (tikz.dev)
YOLO26: YOLO Model for Real-Time Vision AI [2026] (blog.roboflow.com)
06.22 Prompt Injection as Role Confusion (role-confusion.github.io)
Moebius Project Page (hustvl.github.io)
06.21 Inside Anthropic, the $965 Billion AI Juggernaut | The Circuit - YouTube (www.youtube.com)
Building Reliable Agentic AI Systems (martinfowler.com)
Slow breathing impacts inter-organ dynamics modulating brain function and... (www.cell.com)
06.20 Where to Find the Colors Your Screen Can’t Show You – Ryan Moulton's Articles (moultano.wordpress.com)
one-task-per-session rule AI agents - Reddit Search! (www.reddit.com)
Eval harness: What it is, how to use it, and why you should care | DeepEval... (deepeval.com)
Surprising Economics of Load-Balanced Systems - Marc's Blog (brooker.co.za)
06.19 Scaling the humanoid robotics supply chain into billion-dollar wins | McKinsey (www.mckinsey.com)
06.17 Lore | Next-Generation Open Source Version Control - Lore (www.unrealengine.com)
06.16 Running local models is good now | ✰Vicki Boykis✰ (vickiboykis.com)
Making HTTP requests from a container that has no curl, using bash /dev/tcp... (mareksuppa.com)
North Mini Code: Agentic Coding Model for Developers | Cohere (cohere.com)
06.15 Why It’s Nearly Impossible to Build a Robot Without China - The New York Times (archive.is)
06.14 InstaVM | Production Sandbox for AI Agents (instavm.io)
GitHub - gildas-lormeau/SingleFile: Web Extension for saving a faithful copy... (github.com/gildas-lormeau)
Pac-Man, but you're the ghost | Garrit's Notes (garrit.xyz)
Doing nothing at work (www.seangoedecke.com)
06.13 cortecs - Europe's AI Gateway (cortecs.ai)
AI Coding at Home Without Going Broke | Hacker News (news.ycombinator.com)
AI Coding at Home Without Going Broke | Stephen Bochinski (stephen.bochinski.dev)
06.11 GitHub - NousResearch/hermes-agent: The agent that grows with you · GitHub (github.com/NousResearch)
Ear Training (tonedear.com)
Deep Dive: Where Value Accrues in the AI Stack (chamath.substack.com)
06.10 Steve Blank The Mission Model Canvas – An Adapted Business Model Canvas for... (steveblank.com)
06.09 The effective sample size – Alex Smola (alex.smola.org)
OpenCV 5 Deep Dive: A New Foundation for Computer Vision (opencv.org)
06.08 AI is slowing down | Hacker News (news.ycombinator.com)
AI Is Slowing Down (www.wheresyoured.at)
When AI builds itself \ Anthropic (www.anthropic.com)
06.07 Harness engineering: leveraging Codex in an agent-first world | OpenAI (openai.com)
06.06 [2606.05818] Benchmarks in Leipzig (arxiv.org)
Ask HN: What was your "oh shit" moment with GenAI? | Hacker News (news.ycombinator.com)
How LLMs Actually Work | 0xkato (0xkato.xyz)
06.05 GitHub - anthropics/defending-code-reference-harness: Skills for threat... (github.com/anthropics)
GitHub - geo-tp/ESP32-Bit-Pirate: A Hardware Hacking Tool with Web-Based CLI... (github.com/geo-tp)
Ask HN: What is your (AI) dev tech stack / workflow? | Hacker News (news.ycombinator.com)
C++: The Documentary released today – Sutter’s Mill (herbsutter.com)
06.04 Haven Blog: Retro-Tech Parenting (havenweb.org)
A Functional Taxonomy of World Models - Dr. Fei-Fei Li (drfeifei.substack.com)
06.03 Mathematicians issue warning as AI rapidly gains ground | Science | AAAS (www.science.org)
GitHub - pewdiepie-archdaemon/odysseus: Self-hosted AI workspace. · GitHub (github.com/pewdiepie-archdaemon)
05.30 Notes from the AI Now Summit by Mistral | Koen van Gilst (koenvangilst.nl)
You can just say it | caleb gross (noperator.dev)
05.29 We should be more tired than the model | ✰Vicki Boykis✰ (vickiboykis.com)
This AI startup will clean your home for free to train future robots | The Verge (www.theverge.com)
LFM2.5-8B-A1B: an Even Better on-Device Mixture-of-Experts | Liquid AI (www.liquid.ai)
SQLite is All You Need for Durable Workflows - Blog (obeli.sk)
05.28 Elodin - The future of aerospace software (www.elodin.systems)
The lone lisp heap (www.matheusmoreira.com)
Claude Opus 4.8 | Hacker News (news.ycombinator.com)
Generative Recursive Reasoning (arxiv.org)
I think Anthropic and OpenAI have found product-market fit | Hacker News (news.ycombinator.com)
05.27 DeepSeek makes the V4 Pro price discount permanent | Hacker News (news.ycombinator.com)
05.26 DeepSeek-V3 Explained 1: Multi-head Latent Attention | by Shirley Li | TDS... (medium.com)
05.23 AI Engineering from Scratch (aiengineeringfromscratch.com)
Your function's doppelgänger (fedemagnani.github.io)
The quadratic sandwich (fedemagnani.github.io)
05.22 Microsoft starts canceling Claude Code licenses | The Verge (archive.is)
05.20 An OpenAI model has disproved a central conjecture in discrete geometry | OpenAI (openai.com)
05.19 Empirical Research Assistance (ERA): From Nature publication to catalyzing... (research.google)
GitHub - antoinezambelli/forge: A Python framework for self-hosted LLM... (github.com/antoinezambelli)
Mistral AI Acquires Emmi AI to Create the Leading AI Stack | Hacker News (news.ycombinator.com)
05.18 Welcome to Learn Harness Engineering | Learn Harness Engineering (walkinglabs.github.io)
Project Glasswing: what Mythos showed us (blog.cloudflare.com)
Ask an Astronaut (askanastronaut.issinrealtime.org)
05.17 The third hard problem (mmapped.blog)
Accelerando (www.antipope.org)
3D Gaussian Splatting in a Weekend (bfeldman.me)
05.16 Six SQL patterns I use to catch transaction fraud (analytics.fixelsmith.com)
05.15 The Sigmoids Won't Save You - by Scott Alexander (www.astralcodexten.com)
Cut Off - by Anton Leicht - Threading the Needle (writing.antonleicht.me)
05.14 Wie China Markt für Markt erobert (dezernatzukunft.org)
GitHub - DrCatHicks/learning-opportunities: A Claude or Codex skill for... (github.com/DrCatHicks)
Arena AI Model ELO History (mayerwin.github.io)
Introducing Claude for Small Business \ Anthropic (www.anthropic.com)
05.13 The US Is Winning the AI Race (avkcode.github.io)
05.12 GitHub - cactus-compute/needle: 26m function call model that runs on... (github.com/cactus-compute)
On Rendering the Sky, Sunsets, and Planets - The Blog of Maxime Heckel (blog.maximeheckel.com)
GitHub - statewright/statewright: State machine guardrails for AI agents · GitHub (github.com/statewright)
05.11 If AI Writes Your Code, Why Use Python? | by Noah Mitchem | Apr, 2026 | Medium (medium.com)
dBase: 1979-2026 - by Delphi Critic (delphinightmares.substack.com)
Running local models on an M4 with 24GB memory | jola.dev (jola.dev)
05.09 4 Claude Code Subagent Mistakes That Kill Your Workflow (And The Fixes) -... (medium.com)
05.08 GitHub - regent-vcs/re_gent: Git for AI coding agents. · GitHub (github.com/regent-vcs)
05.07 Agents need control flow, not more prompts | Hacker News (news.ycombinator.com)
agents need control flow, not more prompts | brian’s thoughts (bsuh.bearblog.dev)
AlphaEvolve: Gemini-powered coding agent scaling impact across fields —... (deepmind.google)
[2603.30016] Architecting Secure AI Agents: Perspectives on System-Level... (arxiv.org)
A Theory of Deep Learning | Elements of a Vector Space (elonlit.com)
05.06 GitHub - dair-ai/AI-Papers-of-the-Week: 🔥Highlighting the top ML papers... (github.com/dair-ai)
Appearing Productive in The Workplace — No One's Happy (nooneshappy.com)
Wiki Builder: A Claude Code Plugin for Building LLM Knowledge Bases |... (academy.dair.ai)
05.05 When everyone has AI and the company still learns nothing | Hacker News (news.ycombinator.com)
Lessons for Agentic Coding: What should we do when code is cheap? | Hacker News (news.ycombinator.com)
10 Lessons for Agentic Coding (www.dbreunig.com)
[2604.26752] GLM-5V-Turbo: Toward a Native Foundation Model for Multimodal Agents (arxiv.org)
GitHub - angelos-p/llm-from-scratch · GitHub (github.com/angelos-p)
05.03 Rift · For thirty years I programmed with Phish on, every day. In 2026, the... (christophermeiklejohn.com)
The Law of Leaky Abstractions – Joel on Software (www.joelonsoftware.com)
A Couple Million Lines of Haskell: Production Engineering at Mercury | The... (blog.haskell.org)
05.02 DeepSeek V4—almost on the frontier, a fraction of the price (simonwillison.net)
[2509.00462] AI Self-preferencing in Algorithmic Hiring: Empirical Evidence... (arxiv.org)
GitHub - nexu-io/open-design: 🎨 Local-first, open-source alternative to... (github.com/nexu-io)
[2406.11717] Refusal in Language Models Is Mediated by a Single Direction (arxiv.org)
Show HN: AI CAD Harness | Hacker News (news.ycombinator.com)
Install Adam Fusion Extension (fusion.adam.new)
The 2026 Global Intelligence Crisis - Citadel Securities (www.citadelsecurities.com)
04.30 GitHub - NahimNasser/pu · GitHub (github.com/NahimNasser)
I built a Game Boy emulator in F# (nickkossolapov.github.io)
You can beat the binary search – Daniel Lemire's blog (lemire.me)
04.29 Why I Still Reach for Scheme and Lisp Instead of Haskell - jointhefreeworld (jointhefreeworld.org)
04.26 Orinoco: young generation garbage collection · V8 (v8.dev)
Databases Were Not Designed For This (arpitbhayani.me)
From Benchmarketing to Benchmaxxing: What 40 Years of Database Evals Can... (www.typedef.ai)
Introducing mine, a Coalton and Common Lisp IDE | The Coalton Programming Language (coalton-lang.github.io)
GitHub - lucas-maes/le-wm: Official code base for LeWorldModel: Stable... (github.com/lucas-maes)
LeWorldModel: Stable End-to-End Joint-Embedding Predictive Architecture from Pixels (le-wm.github.io)
Turning a Gaussian Splat Into a Videogame | PlayCanvas Blog (blog.playcanvas.com)
Optimizing Datalog for the GPU | Proceedings of the 30th ACM International... (dl.acm.org)
Exposing Floating Point – Bartosz Ciechanowski (ciechanow.ski)
04.25 A 3D Body from Eight Questions — No Photo, No GPU | Clad Blog (clad.you)
04.21 Tim Cook’s Impeccable Timing – Stratechery by Ben Thompson (stratechery.com)
Clojure - Transducers (clojure.org)
04.20 Quantum Computers Are Not a Threat to 128-bit Symmetric Keys (words.filippo.io)
04.19 When moving fast, talking is the first thing to break - daverupert.com (daverupert.com)
The seven programming ur-languages (2022) | Hacker News (news.ycombinator.com)
madhadron - The seven programming ur-languages (madhadron.com)
The Bromine Chokepoint: How Strife in the Middle East Could Halt Production... (warontherocks.com)
04.18 The electromechanical angle computer inside the B-52 bomber's star tracker (www.righto.com)
I'm Coding by Hand - Miguel Conner (miguelconner.substack.com)
Building Effective AI Agents \ Anthropic (www.anthropic.com)
04.17 500 Lines or Less A Python Interpreter Written in Python (aosabook.org)
04.16 Qwen (qwen.ai)
Artifacts: versioned storage that speaks Git (blog.cloudflare.com)
GitHub - exo-explore/exo: Run frontier AI locally. · GitHub (github.com/exo-explore)
04.15 Cybersecurity Looks Like Proof of Work Now (www.dbreunig.com)
Do You Even Need a Database? - DB Pro Blog (www.dbpro.app)
Low-background steel - Wikipedia (en.wikipedia.org)
Want to Write a Compiler? Just Read These Two Papers. (prog21.dadgum.com)
04.14 GitHub - dropseed/plain: The Python web framework for building apps. · GitHub (github.com/dropseed)
Trusted access for the next era of cyber defense | OpenAI (openai.com)
We May Be Living Through the Most Consequential Hundred Days in Cyber... (ringmast4r.substack.com)
Design and Implementation of DuckDB Internals – DuckDB (duckdb.org)
Sometimes powerful people just do dumb shit (www.joanwestenberg.com)
Welcome - GAIA SDK (amd-gaia.ai)
04.13 [2603.12231] Temporal Straightening for Latent Planning (arxiv.org)
04.12 GitHub - smhanov/auth: Boring user authentication code for golang · GitHub (github.com/smhanov)
How I run multiple $10K MRR companies on a $20/month tech stack | Steve Hanov's Blog (stevehanov.ca)
HIPAA compliant AI (www.johndcook.com)
04.11 The Problem That Built an Industry // a.s (ajitem.com)
Center for Responsible, Decentralized Intelligence at Berkeley (rdi.berkeley.edu)
Claude Mythos Preview \ red.anthropic.com (red.anthropic.com)
20 Years on AWS and Never Not My Job (www.daemonology.net)
04.10 From Spark to System (Part I) – THE VOICE IN THE MACHINE (voiceinthemachine.com)
1D-Chess (rowan441.github.io)
llm-wiki · GitHub (gist.github.com)
04.09 A visual introduction to tree calculus | by Johannes Bader | Medium (olydis.medium.com)
Tree Calculus (treecalcul.us)
Research-Driven Agents: What Happens When Your Agent Reads Before It Codes |... (blog.skypilot.co)
04.08 They’re Made Out of Meat - TERRY BISSON of the UNIVERSE (www.terrybisson.com)
https://www.bzarg.com/p/how-a-kalman-filter-works-in-pictures/ (www.bzarg.com)
Kalman Filter Explained Through Examples (kalmanfilter.net)
The Git Commands I Run Before Reading Any Code (piechowski.io)
Porting Mac OS X to the Nintendo Wii | Bryan Keller’s Dev Blog (bryankeller.github.io)
04.07 Google’s timeline for PQC migration (blog.google)
Neha's Writings (nehanarula.org)
Project Glasswing: Securing critical software for the AI era \ Anthropic (www.anthropic.com)
Simplify, Then Add Lightness - by Zack Anderson (blog.zacka.io)
Sam Altman May Control Our Future—Can He Be Trusted? | The New Yorker (archive.ph)
A Cryptography Engineer’s Perspective on Quantum Computing Timelines (words.filippo.io)
04.06 GitHub - fikrikarim/parlor: On-device, real-time multimodal AI. Have natural... (github.com/fikrikarim)
GitHub - arman-bd/guppylm: A ~9M parameter LLM that talks like a small fish. · GitHub (github.com/arman-bd)
[2602.10181] Why do we do astrophysics? (arxiv.org)
04.05 musicForProgramming(); (musicforprogramming.net)
Introducing nanocode: The best Claude Code that $200 can buy. ·... (github.com/salmanmohammadi)
German implementation of eIDAS will require an Apple/Google account to... (news.ycombinator.com)
Baby’s Second Garbage Collector (www.matheusmoreira.com)
Eight years of wanting, three months of building with AI - Lalit Maganti (lalitm.com)
Worse Is Better (dreamsongs.com)
Rise of Worse Is Better (dreamsongs.com)
04.04 A Survey of the 7 Configuration Changes That Turn a Multi-homed Linux Host... (patrickmccanna.net)
Emotion concepts and their function in a large language model \ Anthropic (www.anthropic.com)
04.03 What we learned building 200+ API integrations with OpenCode | Nango Blog (nango.dev)
apfel - Free AI on Your Mac (apfel.franzai.com)
April 2026 TLDR setup for Ollama + Gemma 4 12B on a Mac mini (Apple Silicon)... (gist.github.com)
How Microsoft Vaporized a Trillion Dollars (isolveproblems.substack.com)
04.01 Randomness on Apple Platforms · Luna’s Blog (blog.xoria.org)
PrismML — About (prismml.com)
03.30 Learn Claude Code Interactively — by Ahmed Nagdy (claude.nagdy.me)
03.29 Creating West Coast Buddhism - by Palladium Editors (letter.palladiummag.com)
The rise and fall of IBM's 4 Pi aerospace computers: an illustrated history (www.righto.com)
03.28 RSA and Python | xnacly - blog (xnacly.me)
03.26 The Neo Industrial Age: What Comes After Deep Tech (neoindustrial.substack.com)
Quantization from the ground up | ngrok blog (ngrok.com)
Two studies in compiler optimisations · hmpc (www.hmpcabral.com)
03.25 Long-running Claude for scientific computing \ Anthropic (www.anthropic.com)
03.24 GitHub - t8/hypura: Run models too big for your Mac's memory · GitHub (github.com/t8)
03.21 Some Things Just Take Time | Armin Ronacher's Thoughts and Writings (lucumr.pocoo.org)
A Japanese Glossary of Chopsticks Faux Pas | Nippon.com (www.nippon.com)
03.20 OpenCode | The open source AI coding agent (opencode.ai)
Scaling Karpathy's Autoresearch: What Happens When the Agent Gets a GPU... (blog.skypilot.co)
GitHub - newptcai/purr: A CLI interface for using KittenTTS · GitHub (github.com/newptcai)
03.19 GitHub - VoltAgent/awesome-claude-code-subagents: A collection of 100+... (github.com/VoltAgent)
03.18 open-index/hacker-news · Datasets at Hugging Face (huggingface.co)
Preface - The Emerging Science of Machine Learning Benchmarks (mlbenchmarks.org)
03.17 GitHub - gsd-build/get-shit-done: A light-weight and powerful... (github.com/gsd-build)
03.14 Essays · Gwern.net (gwern.net)
The Optimization Ladder - Cemrehan Çavdar (cemrehancavdar.com)
Emacs and Vim in the Age of AI | (think) (batsov.com)
03.11 LEGO's 0.002mm Specification and It's Implications for Manufacturing -... (www.thewave.engineer)
03.08 Pushing and Pulling: Three Reactivity Algorithms | Jonathan's Blog (jonathan-frere.com)
Why can’t you tune your guitar? – The Ethan Hein Blog (www.ethanhein.com)
GitHub - 0x0mer/CasNum · GitHub (github.com/0x0mer)
Emacs Internal #02: Data First — Deconstructing Lisp_Object in C | The Cloudlet (thecloudlet.github.io)
03.07 Cube 2: Sauerbraten (sauerbraten.org)
The 40x Team | SageOx (sageox.ai)
03.06 Hatchet · Supertoast tables (hatchet.run)
03.04 When AI Writes the World's Software, Who Verifies It? (leodemoura.github.io)
An Interactive Intro to CRDTs (jakelazaroff.com)
VoxRay Games Pushes Major Update To Voxile: "The Builder's Update" (elbowgreasegames.substack.com)
03.02 How I built a sub-500ms latency voice agent from scratch | Nick Tikhonov (www.ntik.me)
03.01 AI Made Writing Code Easier. It Made Engineering Harder. (www.ivanturkovic.com)
Decision trees – the unreasonable power of nested decision rules | Hacker News (news.ycombinator.com)
Decision Trees (mlu-explain.github.io)
microgpt (karpathy.github.io)
02.28 Code Mode: give agents an entire API in 1,000 tokens (blog.cloudflare.com)
Stop Burning Your Context Window — We Built Context Mode | Mert Köseoğlu (mksg.lu)
Cognitive Debt: When Velocity Exceeds Comprehension | rockoder (www.rockoder.com)
[2602.21472] The Design Space of Tri-Modal Masked Diffusion Models (arxiv.org)
The Future of AI | Coffee and PI (lucijagregov.com)
The Eternal Promise: A History of Attempts to Eliminate Programmers (www.ivanturkovic.com)
02.27 microcodeagent · GitHub (gist.github.com)
02.26 Steering Interpretable Language Models (www.guidelabs.ai)
How will OpenAI compete? — Benedict Evans (www.ben-evans.com)
02.25 The Blade Runner Unicorn - GeekyDad Crafts (geekydadcrafts.com)
02.22 THE 2028 GLOBAL INTELLIGENCE CRISIS (www.citriniresearch.com)
How Taalas “prints” LLM onto a chip? | Hacker News (news.ycombinator.com)
How Taalas "prints" LLM onto a chip? (www.anuragk.com)
How I Use Claude Code | Boris Tane (boristane.com)
02.21 Huy Vector DIY (www.huyvector.org)
Full Documentation (github.com/tnm)
How Tinfoil Proves Exactly What Model Is Running (tinfoil.sh)
02.17 Aeris — Real-Time 3D Flight Tracking (aeris-flight.vercel.app)
02.16 The Economics of LLM Inference: Batch Sizes, Latency Tiers, and Why Model... (mlechner.substack.com)
02.15 Arming the rebels with GPUs: Gradium, Kyutai, and Audio AI | Amplify Partners (amplifypartners.com)
Knowledge distillation - Wikipedia (en.wikipedia.org)
Two different tricks for fast LLM inference (www.seangoedecke.com)
Fragments: February 13 (martinfowler.com)
02.14 mickamy/sql-tap: Watch SQL traffic in real-time with a TUI (github.com/mickamy)
02.11 Should Your Developer Company Go Open-Source? (extremefoundership.substack.com)
GLM-OCR/examples/mlx-deploy/README.md at main · zai-org/GLM-OCR (github.com/zai-org)
oss-sec: GNU InetUtils Security Advisory: remote authentication by-pass in telnetd (seclists.org)
The Feynman Lectures on Physics (www.feynmanlectures.caltech.edu)
02.10 Frontier AI agents violate ethical constraints 30–50% of time, pressured by... (news.ycombinator.com)
[2512.20798] A Benchmark for Evaluating Outcome-Driven Constraint Violations... (arxiv.org)
02.08 The Toughest Sell: A Founder's Guide to Startup Exits | Derek Yan (derekyan.com)
Claude's Constitution \ Anthropic (www.anthropic.com)
02.07 Hoot: Scheme on WebAssembly — Spritely Institute (www.spritely.institute)
AI ‘Godfather’ Yann LeCun: LLMs Are Nearing the End, but Better AI Is Coming... (www.newsweek.com)
Rectified LpJEPA | Yilun Kuang (yilunkuang.github.io)
2748 Robust and Interactable World Models in Computer Vision - YouTube (www.youtube.com)
02.06 Claude Composer | Just Joshing (www.josh.ing)
How to effectively write quality code with AI (heidenstedt.org)
The Waymo World Model: A New Frontier For Autonomous Driving Simulation (waymo.com)
The Programmer's Paradox: Systems Thinking (theprogrammersparadox.blogspot.com)
02.04 Building a 24-bit Arcade CRT Display Adapter, From Scratch | Stephen's Site (www.scd31.com)
02.02 Bahtinov mask - Wikipedia (en.wikipedia.org)
02.01 [2512.08296] Towards a Science of Scaling Agent Systems (arxiv.org)
Towards a science of scaling agent systems: When and why agent systems work (research.google)
Apple1 Project - Gallery (apple1.chez.com)
01.31 https://help.apple.com/pdf/security/en_US/apple-platform-security-guide.pdf (help.apple.com)
CPython Internals Explained | Hacker News (news.ycombinator.com)
Euro firms must ditch Uncle Sam's clouds and go EU-native | Hacker News (news.ycombinator.com)
Euro firms must ditch Uncle Sam's clouds and go EU-native • The Register (www.theregister.com)
01.30 moltbook - the front page of the agent internet (www.moltbook.com)
01.28 Doing the thing is doing the thing (www.softwaredesign.ing)
Andrej Karpathy on X: "A few random notes from claude coding quite a bit... (x.com)
Sacred Geometry:Golden Ratio:Equil.Triangle in circle:GeometryCode.com (geometrycode.com)
Prism | Hacker News (news.ycombinator.com)
01.27 Management as AI superpower - by Ethan Mollick (www.oneusefulthing.org)
01.26 Show HN: A small programming language where everything is pass-by-value | Hacker News (news.ycombinator.com)
01.25 Clarity-1: What Worked, and Where We Go Next (albedo.com)
[2601.05047] Challenges and Research Directions for Large Language Model... (arxiv.org)
AI + Human Agent to get tasks done (tendem.ai)
01.24 GPTZero finds 100 new hallucinations in NeurIPS 2025 accepted papers (gptzero.me)
Continuous Autoregressive Language Models | Chenze Shao (shaochenze.github.io)
Yann Le Cun: 'Why I'm leaving Meta to launch my own AI start-up' (archive.is)
Many Small Queries Are Efficient In SQLite (www.sqlite.org)
some C habits I employ for the modern day | ~yosh (www.unix.dog)
Parse, don’t validate (lexi-lambda.github.io)
Unrolling the Codex agent loop | OpenAI (openai.com)
01.23 Your App Subscription Is Now My Weekend Project · Roberto Selbach (rselbach.com)
Bugs Apple Loves (www.bugsappleloves.com)
How Company Culture Shapes Employee Motivation (archive.is)
Recent discoveries on the acquisition of the highest levels of human... (www.science.org)
01.22 The Future of Team Topologies: When AI Agents Dominate — Team Topologies -... (teamtopologies.com)
01.21 The Agentic AI Handbook: Production-Ready Patterns - Log - nibzard (www.nibzard.com)
Are arrays functions? (futhark-lang.org)
01.20 Nvidia Stock Crash Prediction (entropicthoughts.com)
The assistant axis: situating and stabilizing the character of large... (www.anthropic.com)
01.18 ASCII characters are not pixels: a deep dive into ASCII rendering (alexharri.com)
01.17 The Five Orders of Ignorance – Communications of the ACM (cacm.acm.org)
Structured LLM outputs (nanonets.com)
Die AI FIRST AI Assistenten Datenbank – AI FIRST (ai-first.ai)
Legal - Training Data - Datasets used for Apple Intelligence - Apple (www.apple.com)
01.16 Open Policy Agent - Homepage | Open Policy Agent (openpolicyagent.org)
OpenBSD-current now runs as guest under Apple Hypervisor (www.undeadly.org)
How to Instantly Render Real-World Scenes in Interactive Simulation | NVIDIA... (developer.nvidia.com)
COLMAP — COLMAP 3.14.0.dev0 | 5b9a079a (2025-11-14) documentation (colmap.github.io)
01.15 [2601.05637] GenCtrl -- A Formal Controllability Toolkit for Generative Models (arxiv.org)
Daniel's Blog (danilafe.com)
Sparrow-1: Human-Level Conversational Timing in Real-Time Voice (www.tavus.io)
01.12 DeepSeek's mHC: When Residual Connections Explode - Taylor Kolasinski (taylorkolasinski.com)
01.11 Ex-Apple AI Infra Strategy Head & Amplift CEO: Surviving the Shift from... (www.youtube.com)
01.08 [2512.24617] Dynamic Large Concept Models: Latent Reasoning in an Adaptive... (arxiv.org)
The Q, K, V Matrices (arpitbhayani.me)
2026-01-01_hashmap-merge.md (gist.github.com)
A4 Paper Stories - Susam Pal (susam.net)
01.07 OHGOD Philosophy | One Hobbyist GPU, One Day (ohgodmodels.xyz)
01.06 rberg27/doom-coding: A guide for how to use your smartphone to code anywhere... (github.com/rberg27)
.:: Phrack Magazine ::. (phrack.org)
01.05 Faster and enjoyable ZSH (maybe) (htr3n.github.io)
So You Think You've Awoken ChatGPT — LessWrong (www.lesswrong.com)
01.04 Neural Networks: Zero To Hero (karpathy.ai)
01.01 2025 letter | Dan Wang (danwang.co)
Prompt für Character Sheet (marketing-ki.de)