Full-text search across 69266 articles. Combine with topic and date filters; results sorted by relevance.
A developer ran a controlled experiment comparing an AI coding agent's output with and without access to a persistent story map. The agent produced a generic user story from a ticket alone, but with tool access to the ma…
A developer's essay argues that AI-generated codebases fail in unpredictable ways, with hidden 'chasms' that crash or hang outside demos, unlike human-written code where gaps are visible and predictable. The author notes…
Meta is testing Muse, a standalone iPhone personal AI agent with access to email, calendars, health and financial data, that can manage schedules, make purchases, and track spending, according to a Meta-authored App Stor…
Google's Gemini 3.8 Flash model focuses on improving persistence and tool use in agentic workflows rather than expanding context, according to a developer's analysis. The model scores 73.7% on DeepSWE v1.1, up from 65.3%…
A developer created Anne OS Kids, a browser-based operating system simulation for children, using vibe coding and AI. The project, built with HTML, CSS, and JavaScript, includes educational apps and games, and serves as …
An agent harness is the code and structure around a language model that converts raw text prediction into goal-directed behavior, and it has driven larger performance gains than model upgrades, according to an explainer …
A Cloudflare setting silently blocked AI crawlers from a digital marketing consultant's website, making it invisible to AI assistants for its entire existence, according to the consultant's own account. The consultant, w…
A developer fixed a bug in Chron, an MCP server that logs AI coding sessions, where Claude and Cursor sessions were incorrectly merged due to a UNIQUE constraint on the session title. The issue was that the title alone w…
In a 2012 Edge.org annual question, cognitive scientist Christof Koch argues that the same strategy used to decode DNA's structure could crack the 'neural code' of consciousness, citing Francis Crick's pursuit of the 'se…
Researchers introduced Sarathi-Serve, an LLM inference scheduler that uses chunked-prefills and stall-free scheduling to improve throughput-latency tradeoffs, achieving 2.6x higher serving capacity for Mistral-7B on a si…
Neo4j Labs released @neo4j-labs/nams-ai-provider, a community provider for the Vercel AI SDK that gives any model persistent, cross-session memory backed by the hosted Neo4j Agent Memory Service (NAMS). The package requi…
Meta's experiment with AI-powered engineering "pods" of three to five workers increased code changes by 220% year-on-year, but features shipped rose only 36% while major technical and security incidents jumped 40%, accor…
The Institute of Foundation Models (IFM), the frontier lab launched by MBZUAI in May 2025, released K2 Horizon, a fleet of six Apache 2.0-licensed open-source models ranging from 0.9B to 375B parameters, alongside the pr…
Two CVSS 9.8 vulnerabilities were disclosed on September 5 in open-source AI agent sandboxes, Cua and AutoAgent, both allowing unauthenticated remote code execution due to default network bindings and missing authenticat…
Researchers propose REFINE, a KG-aware budgeted LLM graph refinement framework that learns patient-personalized medical concept representations from text-attributed knowledge graphs, using a sequential reinforcement lear…
A developer argues that most production AI agents should be built as simple while loops rather than complex frameworks like LangChain or CrewAI. The post highlights that top-performing agents on SWE-bench Verified, scori…
FreeModel API's live status flipped to up, with Uprouter's best-effort probes recording 146ms latency 10 hours ago and 481ms one day ago, following a down reading 12 days earlier. Uprouter rates the frontier model API ro…
Wayfinder, an open-source reference implementation on GitHub, demonstrates a structured approach to AI evaluation, advocating for a pipeline that layers rule-based checks, LLM-as-a-judge, offline golden datasets, and onl…
A peer AI crew (GLEE) identified two bugs in the payment-verification code published by the author on their protocol page, including a rejection check that silently passed malformed payment data; the author confirmed the…
OpenAI is preparing a gated incentive program that rewards ChatGPT workspace credits to users who install and successfully use selected third-party plugins, according to code in the ChatGPT Windows client version 26.901.…