February 2026 Archive
12001.
I built a hybrid Fuzzy/LLM pipeline to rank knife steels from Reddit (new.knife.day)
12002.
Text of Terry Pratchett's 1995 Bill Gates Interview (old.reddit.com)
12003.
Show HN: 2026 US take-home pay calculator (updated tax brackets) (taxcalc2026.xyz)
12004.
Show HN: RemoveTheBG – Free AI background remover for ecommerce photos (removethebg.com)
12005.
Why Normativity Remains Challenging for LLMs (link.springer.com)
12006.
Dare Mighty Code (nasa-jpl.github.io)
12007.
Show HN: Ruby-libgd – Native 2D graphics engine for Ruby (no CLI, no JavaScript)
12008.
Graph-based multi-agents smash long-context benchmarks–89% MMLU-Pro on 8B models (github.com)
12009.
Show HN: I built OLTP in my memory system. Looking for opinions
12010.
I made a small tool that lets you (partially) override ChatGPT's Systemprompt (chromewebstore.google.com)
12011.
Hyperstar: LiveView for TS/JSX (github.com)
12012.
We stopped paying for certifications and built an internal "Lab" instead (one2n.io)
12013.
BitBills Zero-Day (non-destructive private key sweeping) (bitcointalk.org)
12014.
An offline, in-memory file conversion workbench for regulated environments (factual7418.gumroad.com)
12015.
Show HN: API router that picks the cheapest model that fits each query (komilion.com)
12016.
Show HN: Agent Audit – Open-source security scanner for AI agents (github.com)
12017.
Context Engineering for Coding Agents (martinfowler.com)
12018.
Thoughts on the job market in the age of LLMs (interconnects.ai)
12019.
Nvidia commits 3x more code across 30k developers with Cursor (cursor.com)
12020.
Steam Machine and Steam Frame delays are the latest product of the RAM crisis (arstechnica.com)
12021.
Why it's so hard to fuel the Artemis rockets (nationalgeographic.com)
12022.
Tree hug dream unlocked by AI (tiago.rio.br)
12023.
Control Layer for AI (blog.dottxt.ai)
12024.
Live crypto treasury valuation: modeling MSTR and BMNR with a "5% alchemy (pilote.fund)
12025.
Show HN: Vibe-coded game prototypes. Tell me which to work on
12026.
My Solution to LeetCode Interviews (entrevue.app)
12027.
HelloRL: A modular framework, like Lego for Reinforcement Learning (github.com)
12028.
The Jurassic Debate: AI (arturonereu.com)
12029.
The Agentic Inversion Principle (mikebrevoort.com)
12030.
No Mac mini nor coding needed, Cloudflare OpenClaw Wrapper (moltworker-cloud.pages.dev)