Kotlin creator’s new language: a formal way to talk to LLMs instead of English

kotlin-creator’s-new-language:-a-formal-way-to-talk-to-llms-instead-of-english

Case Study Code LOC[1] Spec LOC[1] Shrink Factor Tests Passed WebVTT subtitles support for yt-dlp (video downloader) 255 38 6.7x before: 1241/1242 after: 1278/1279 (37 tests added) Italian SSN generator for Faker (python library for generating mock data) 165[2] 21 7.9x before: 2216 after: 2229 (13 tests added) Encoding auto-detection and normalization for beautifulsoup4 (Python […]

Iranian Hacktivists Strike Medical Device Maker Stryker and Wiped Systems

iranian-hacktivists-strike-medical-device-maker-stryker-and-wiped-systems

Stryker, a leading maker of medical devices, was hit early this morning with a cyberattack that has reportedly caused the company’s systems to shut down globally. The company has acknowledged the attack and called it ”severe” in communication with employees. A known Iranian hacktivist group named Handala posted messages on hacked systems and on social […]

Show HN: Axe A 12MB binary that replaces your AI framework

show-hn:-axe-a-12mb-binary-that-replaces-your-ai-framework

A CLI tool for managing and running LLM-powered agents. Most AI tooling assumes you want a chatbot. A long-running session with a massive context window doing everything at once. But that’s not how good software works. Good software is small, focused, and composable. Axe treats LLM agents the same way Unix treats programs. Each agent […]

Malus – Clean Room as a Service

Robot-Powered Clean Room Recreation Our proprietary AI systems have never seen the original source code. They independently analyze documentation, API specifications, and public interfaces to recreate functionally equivalent software from scratch. The result is legally distinct code that you own outright. No derivative works. No license inheritance. No obligations. ✓ 100% robot-written code ✓ Zero […]

Show HN: We analyzed 1,573 Claude Code sessions to see how AI agents work

show-hn:-we-analyzed-1,573-claude-code-sessions-to-see-how-ai-agents-work

Try the hosted version for free at rudel.ai Analytics for Claude Code. Rudel gives you a dashboard with insights on your coding sessions — token usage, session duration, activity patterns, model usage, and more. Prerequisites Bun runtime installed Getting Started Create an account at app.rudel.ai Install the CLI and connect it to your account: npm […]

Document poisoning in RAG systems: How attackers corrupt AI’s sources

document-poisoning-in-rag-systems:-how-attackers-corrupt-ai’s-sources

I injected three fabricated documents into a ChromaDB knowledge base. Here’s what the LLM said next. In under three minutes, on a MacBook Pro, with no GPU, no cloud, and no jailbreak, I had a RAG system confidently reporting that a company’s Q4 2025 revenue was $8.3M, down 47% year-over-year, with a workforce reduction plan […]

Electric motor scaling laws and inertia in robot actuators

electric-motor-scaling-laws-and-inertia-in-robot-actuators

This is the beginning of a series of posts about robot actuation. The intent here is not to prescribe some specific architecture or solution, but to talk about the fundamentals, with as little bias as I can manage. Hopefully you learn something. Here’s an ostensibly straightforward question for anyone who’s spent time thinking about robot […]

Slicing Bezier Surfaces

slicing-bezier-surfaces

It is possible to split a bezier surface into two pieces. It helps us to slice a surface into several pieces without losing the original shape. In that way we are able to delete one piece of several slices of a whole surface. function splitBezier(t, p0, p1, p2, p3) { const p4 = lerp(t, p0, […]

Are LLMs not getting better?

are-llms-not-getting-better?

I was reading the metr article on how llm code passes test much more often than it is of mergeable quality. They look at the performance of llms doing programming when the success criterion is “passes all tests” and compare it to when the success criterion is “would get approved by the maintainer”. Unsurprisingly, llm […]