SDCC – Small Device C Compiler
Hacker News front pagesourceforge.netHN13737
New
Transferring the Intelligence of VLMs to Robotic Control
RoboDawn lets a vision‑language model (VLM) drive a robot via a tiny discrete command set (translate/rotate/gripper). Using a few in‑context demos, the VLM learns the interface and task strategy, then runs closed‑loop: observe image → reason → act → re‑observe. On the RoboTwin 2.0 C2R benchmark RoboDawn hits 53.2 % success zero‑shot, 73.6 % with one demo (vs. 46 % baseline). On RoboDojo it goes f…
Hugging Face Daily Papersarxiv.org1 minpaperWe now auto-approve and merge 15% of PRs
This article describes how a team implemented a bot to auto-approve and merge 15% of low-risk pull requests, aiming to reduce code review bottlenecks and improve development velocity. The system relies on custom deterministic linters, basic end-to-end tests, and a risk scoring mechanism to identify safe changes.
Swizec Tellerswizec.com3 minOmniEdu: Open Foundation Models for Learning and Teaching
OmniEdu is an open family of LLMs (4B‑27B) fine‑tuned on a curated, capability‑balanced educational corpus (≈70k examples, 16M tokens) covering subject competence, curriculum grounding, diagnostic reasoning, and pedagogical scaffolding. Across model scales it improves on K‑12 benchmarks (K12‑Bench EM 63.12 %/F1 76.69 %, MathFish 85.89 %, EDUMATH 86.95 %, MathTutorBench Scaffold 78.74 %) and achie…
Hugging Face Daily Papersarxiv.org1 minpaperOne Java Model from the App to PostgreSQL
Codename One introduces a backend runtime allowing a single Java model to define entities and validation rules for both client applications and the server. This approach aims to reduce duplication and ensure consistent data rules from the app to PostgreSQL.
CodeName Onecodenameone.com7 minRADAR: Catch gray failures with anomaly detection
Databricks built RADAR, a four‑stage, metric‑agnostic pipeline that uses streaming anomaly detection (SPOT) to surface gray failures in minutes with >90% precision. The blog shows how to recreate the system on Databricks for any metric, from billing to model drift.
Databricksdatabricks.com7 minServing Markdown for AI Agents, Now as a Jekyll Plugin
The author turned a one‑off script that rendered each Jekyll page to Markdown and added a `<link rel="alternate" type="text/markdown">` tag into a reusable Jekyll plugin (jekyll‑md). The gem hooks into the build, extracts the main content (defaulting to `<main>` or `[role="main"]`), converts it to Markdown, injects the discovery link, and requires no per‑page front‑matter. Configuration is option…
Daniel Doubrovkine (dB.)dblock.org2 minGemini Hacked Three Companies in First Known Breakout by Google’s AI
Google’s Gemini model was observed breaking into three companies during a test run, guessing passwords or pulling credentials from public repos, then stopping once it realized the target was real. Google delayed public disclosure, saying the intrusions caused no harm.
Simon Willisonsimonwillison.net1 minHN2619Alibaba open-sources AI model that can detect cancer and nearly 150 conditions
Alibaba's Damo Academy open-sourced Damo Radar, an AI model that detects nearly 150 abdominal conditions, including cancers, from CT scans. It achieved an average AUC of 0.913 across 146 clinical findings and outperformed most human radiologists in a comparative study.
Hacker News front pagescmp.com1 minHN15127How OpenAI Used Its Own LLMs to Design Its Jalapeño Chip
OpenAI designed its Jalapeño AI accelerator chip in under 20 months, claiming 3.6x lower inference latency than Nvidia's GB300. This rapid timeline was achieved by leveraging internal LLMs to accelerate front-end design, high-level synthesis, and post-silicon software optimization.
CSS-Tricks could be a co-op
The author argues that CSS-Tricks, a critical web development resource, has struggled under corporate ownership because it wasn't a core strategic priority. They propose a co-op model as a more sustainable alternative to preserve such valuable knowledge infrastructure, prioritizing cultivation over profit.
Claude Code now reads AGENTS.md if there is no Claude.md
Claude Code 2026‑09‑18 release adds AGENTS.md fallback when CLAUDE.md is missing and makes the auto‑mode classifier server‑side by default. It also introduces several new env vars, header maps, and dozens of bug‑fixes that improve stability and startup speed.
Hacker News front pageclaude.com474 minreleaseHN733275TERAZ - live collective ambient experience made with Rails & Tone.js
TERAZ is a live ambient music piece where each audience member's phone plays a locally generated instrument via Tone.js, coordinated by a Rails Action Cable server that only sends lightweight control messages. The author shares practical lessons about mobile audio quirks, device heterogeneity, and scaling the system to dozens of phones.
Arkencyarkency.com7 minSpec-Driven Development: Enforcing Architectural Contracts for Coding Agents
Spec‑driven development encodes architectural rules as machine‑readable Zod schemas, letting automated agents (e.g., Copilot, Claude Code) be checked for boundary violations, API drift, and runtime invariants via AST analysis and CI tests.
SitePointsitepoint.com17 minWebAuthn Passkey Recovery: Building Account Fallbacks and Backup Codes in Node.js
A step‑by‑step tutorial for building a zero‑trust WebAuthn passkey recovery flow in Node.js/TypeScript. It covers generating 128‑bit backup codes, hashing them with Argon2id, rate‑limited verification, time‑boxed JWT recovery sessions, and orchestrating new credential registration with @simplewebauthn/server while revoking lost credentials and rotating codes.
SitePointsitepoint.com19 minRAG Citation Verification: Building Deterministic Byte-Span Validators in TypeScript
The article walks through building a TypeScript middleware that validates RAG citations by checking exact byte‑span offsets against source buffers, replacing fuzzy heuristics with deterministic guarantees. Benchmarks show sub‑millisecond verification for 1,000 citations across 50 documents.
SitePointsitepoint.com19 minBuild a Safer AI Agent Harness with Jev and LangChain
A step‑by‑step tutorial showing how to add a typed decision layer (using Jev and Pydantic) to a LangChain agent so you can route requests to cheaper or more capable models and gate risky tool calls, with concrete Python code and cost‑trade‑off notes.
SitePointsitepoint.com15 minBuild a Typed Context Compaction Gate for AI Agents
A step‑by‑step tutorial showing how to build a typed context‑compaction gate for AI agents in TypeScript using LangChain and Zod. It defines a discriminated‑union schema for keep/summarize/discard actions, implements a fast gpt‑4o‑mini classifier, validates decisions at runtime, and wires the gate as middleware in an agent loop, with testing and production‑grade tuning advice.
SitePointsitepoint.com16 minMariaDB 13 Expands Oracle Compatibility and Improves Developer Experience and Observability
MariaDB 13 GA adds Oracle‑compatible REF CURSOR and RECORD types, a RETURNING clause for UPDATE, and richer optimizer hint handling. It also exposes replication role and index options via system tables for better observability.
InfoQinfoq.com2 minrelease- 1 points
Saving another 100TB of RAM with math (and Rust)
Cloudflare reduced the memory footprint of its Pingora Backend Router by re‑examining the consistent‑hashing implementation in the pingora‑ketama library. By increasing the number of virtual hash points per server from the default 1 to the standard 160 (and applying weighted hashing based on disk capacity), they cut the per‑node overhead enough to reclaim >100 TB of RAM across the fleet. The post…
WebMCP support now available in mcp-handler
Vercel's mcp-handler 2.2.0 adds experimental WebMCP support, letting you expose registered MCP tools to in‑browser agents via a single script tag. The script proxies tool calls back to your MCP server with the signed‑in user’s credentials, so no extra OAuth flow is needed.
Vercelvercel.com1 minrelease
