Chess Atlas
chess-timeline.vercel.app
Feeds for You
Show HN: Cactus Needle 3: 8-29MB automation models can match DeepSeek V4 Flash
cactuscompute.comShapelearn Qwen 3.8 27B (13.1 GB VRAM)
byteshape.comBrood War Bench
bw.swerdlow.devHow to talk about "AI" without adding to the anthropomorphization
buttondown.comSpymarks, not Watermarks
brand.ioNamed and Optional Arguments are Awesome
botahamec.devI am often wrong
borischerny.comAMD's random number generator can't generate a 0?
board.flatassembler.netHellGates, custom CPU gate-level challenge
blog.xutaxkamay.comDesign your programming languages right (2024)
blog.veritates.loveSAML: A fractal of bad design
blog.trailofbits.comGitHub Actions leaking secrets when Miri output is cached
blog.rust-lang.orgThe First Microbe for Mars
blog.pioneer-labs.orgBuild Faster Feedback Loops Using Qualitative User Research
blog.nseldeib.comDon't Make Job Referrals Public
blog.melashri.netBend 2 and the Vibe-Coding Trap
blog.liampwll.comA study of sequence weighting at scale
blog.janestreet.comThe new CC, an AI agent built for families
blog.googleInside ZCode: Silently uploading your Git history to the cloud
blog.ferstar.orgSoftware sandboxing: The basics (2025)
blog.emilua.orgThe trouble with 'ntile()'
blog.djnavarro.netTowards Self-Driving Codebases
blog.detail.devI don't want to read what you didn't write
blog.colinbreck.comSaving another 100TB of RAM
blog.cloudflare.comPython Workers are now generally available
blog.cloudflare.comThe UV index is not the warm sensation of sunlight on bare skin
blog.asciitweezers.comWhat Zig felt like, coming from Rust
besok.github.ioBend – a language that blocks AI mistakes via proof and runs on GPUs
bend-lang.comShow HN: JevBench, a reproducible benchmark for typed decision models
benchmarkheaven.comBeanShell3 in Development
beanshell.github.ioWhat Sun got wrong
bcantrill.dtrace.orgBtrfs/ZFS/bcachefs under workloads classic benchmarks skip
bartosz.fenski.plAugustofaces: Pareidolia Fine Art
augusto.atDoes an open-weight decision model beat a hosted one? Jev vs. Laya
astgl.comCustom home server built from spare parts
asmat.caAsk A Monk – A digital wilderness for thoughts with no immediate answer
askamonk.onlineDeepSeek Elastic Compute (DSec): Sandbox Infrastructure for Effective Agentic Training at Scale
arxiv.orgEt Tu, MacBook? Unprivileged Keystroke Inference via Built-In IMU Side Channel
arxiv.orgAn empirical study of harness design for coding agents
arxiv.orgInfinite-Parameter LLMs: Generating and Adapting Weights from Live Data
arxiv.orgShow HN: Training a model to identify AI web content from structure alone
arxiv.orgThe Implications of Linguistic Illegibility for LLM Security
arxiv.orgCompiler-style optimization for drawing via Skia
arxiv.orgCache-to-Cache: Direct Semantic Communication Between LLMs (2025)
arxiv.orgStepfun Step 5 Preview (LLM): On AA Pareto frontier
artificialanalysis.aiMiMo-v2.6-Pro: Intelligence, Performance and Price Analysis
artificialanalysis.aiGPT-6 Sol (Max) Intelligence, Performance and Price Analysis
artificialanalysis.aiClaude Opus 5.5 Intelligence, Performance and Price Analysis (Max)
artificialanalysis.aiTrump's sneaky plan to gut Endangered Species Act draws from 1995 Scalia dissent
arstechnica.com