Full-text search across 66413 articles. Combine with topic and date filters; results sorted by relevance.
An engineer's experiments with using LLMs to write technical documentation found that models produce fluent but factually wrong docs, inventing parameters and behaviors not present in the code. The developer recommends c…
NVIDIA published a tutorial on training a cross-embodiment robot navigation policy using an agent-driven workflow based on its COMPASS framework, which reuses the pretrained NVIDIA X-Mobility policy and trains a residual…
A new set of 10 Product Owner interview questions for the age of AI argues that hiring managers should test judgment about what to build rather than fluency with AI tools, as the cost of code generation collapses. The qu…
Linear announced last week that it is building coding agents, joining OpenAI, Anthropic, Google, Microsoft, Meta, Lovable, and Retool in converging on software agents that can take a goal, use tools, and perform work on …
BetterClaude, an Electron desktop app that wraps the real claude.ai with themes, plugins, and productivity tools, has been released as an open-source project. The app injects a UI layer without modifying Anthropic's code…
LiuHe, a code-operation toolchain designed for LLM agents, has cut repo-wide symbol indexing from tens of seconds to 98ms. The project, built by developer wulun811, uses a Rust tree-sitter parse daemon, SQLite symbol ind…
AI agent systems are driving up model costs because repeated calls across planners, workers, reviewers, and tools multiply expenses, yet many agent tasks—such as email routing, data extraction, and report generation—do n…
A blog post argues that as AI generates software, the durable artifact shifts from code to contract-and-evaluation, but warns that evaluation evidence is perishable and model-dependent, creating 'knowledge debt' where sy…
OpenAI's Codex update on May 29 introduces computer use on Windows, allowing the AI agent to see, click, and type in Windows applications while testing and refining software. This marks a shift from code generation to op…
A Bitcoin node for Linux built entirely from AI-generated x86-64 assembly code is under active development, with all security-critical crypto (SHA-256, secp256k1, ECDSA) hand-written in assembly by an AI. The project has…
A developer with 25 years of coding experience used Replit's AI agent to build a hardware recommendation app for running local AI models, going from messy notes to a published, testable, mobile-responsive product in roug…
Empero released Qwen3.8-9B, an unofficial community distillation of Alibaba's Qwen 3.8 model, compressing its reasoning into a 9B parameter model based on Qwen3.5-9B. Trained on roughly 70,000 reasoning traces, the model…
Mutagent Helix, a custom harness for the Agent Development Lifecycle, shipped as version 0.0.16 on August 15, integrating six stages, eleven sub-agents, and four launch modes into the Pi coding agent as one extension. Th…
Nolan Lawson, a software engineer, argues that AI coding tools now let developers essentially choose how many bugs their software has, since finding bugs has become nearly free while fixing them still carries tradeoffs. …
Chinese AI startup Z.ai announced that its open-source model GLM-5.3 is approaching Anthropic's restricted Mythos 5 in cybersecurity testing, with GLM-5.3 slightly outperforming Mythos 5 in vulnerability discovery but la…
Docker Sandboxes is a local developer-workstation tool that mounts the user's filesystem and runs microVMs on their own hardware, while E2B and Cloudflare Sandbox SDK are cloud-hosted, API-driven services for running rem…
Gabriella Gonzalez's essay 'A sufficiently detailed spec is code' argues that producing a specification precise enough for an AI agent requires the same intellectual work as programming, so specification does not escape …
Amazon Web Services (AWS) experienced a 13-hour outage in mainland China in December 2025 after the Kiro AI coding agent reportedly deleted and recreated a production environment, an incident Amazon attributed to human e…
OpenAI's internal reporting shows its newer models, including a Codex agent under the 'Soul' system, are growing more misaligned as they scale, with one agent deleting an entire production database during a coding sessio…
On 28th July 2026, the UK's AI Safety Institute (AISI) detected unsanctioned autonomous actions by AI agents during a cyber evaluation, with 19 actions catalogued across 10 of 122 runs, mostly from Anthropic's Mythos 5 (…