Full-text search across 67685 articles. Combine with topic and date filters; results sorted by relevance.
Alibaba's Qwen research lab released Qwen 3.8 27B, an open-weights, Apache 2.0-licensed vision-language model that runs entirely on a single consumer workstation with 24 GB GPUs or Apple Silicon Macs. Its GGUF builds top…
Maxio CEO Branden Jenkins discovered his AI agent had burned $1,000 in tokens during a weekend coding session, charged automatically in $1,000 increments to an auto-renewing card. Jenkins said the surprise invoice is les…
A developer's evaluation of twelve AI communities found that smaller servers like PromptCube Discord (3,400 members) and LLM Builders (1,100 members) outperform larger ones in technical message density, response time, an…
Massdriver's blog post argues that citizen developers—employees outside IT who build software—have existed since Lotus 1-2-3 and are now more prevalent than ever, with Gartner finding in 2021 that 41% of employees create…
XAI's Grok 4.6 is now available in GitHub Copilot across eight development surfaces, including VS Code, Visual Studio, and JetBrains IDEs, with pricing at $2 per million input tokens and $6 per million output tokens, rou…
A developer built mcptoon, a zero-dependency CLI tool that reduces AI agent context window usage by keeping MCP tool schemas out of context. The project, which uses only Python's standard library, was motivated by the uv…
An anonymous frontier-class model called Ox Alpha appeared on OpenRouter on August 20, 2026, at 20:04:55 UTC, offering a 1,048,576-token context window, text, image and video input, and zero-priced prompts and completion…
A developer built Infra Lang, an infrastructure-as-code DSL that compiles a single service definition into Kubernetes manifests, Docker Compose, Helm charts, Terraform HCL, and GitHub Actions workflows, with an LSP and l…
Netflix engineer Chris Dollin describes a technique for performing large-scale code changes (LSCs) by using AI agents to explore the problem space and encode solutions into deterministic metaprogramming programs, rather …
Notula, a free desktop app for macOS and Windows, lets humans edit Markdown documents that AI agents read and write, keeping files in the repository with no server or account. The app preserves 96.4% of blocks untouched …
Yubico, a security key maker, reported that approximately 46% of AI-flagged security findings were downgraded during human triage, most commonly from High to Medium severity, based on agentic assessments of 29 open sourc…
Software teams of 5-10 developers can now run 20 to 100+ coding agents in parallel, turning a busy day of 50 commits/20 pushes/10 PRs into 500 commits/200 pushes/100 PRs, according to an analysis of Uber's microservices …
Only a small percentage of generative AI pilots reach production, and the gap is typically in the five layers underneath the model, according to a new article. The piece highlights five tools—LangGraph, E2B, and three ot…
A developer built Relay, a remote control tool for AI agents running on a home server, enabling live terminal access from any device with session persistence, file transfer, and token quota tracking. The open-source Go a…
GitLab, the closest feature-for-feature GitHub alternative, offers a full DevOps lifecycle with built-in CI/CD and security scanning, but its Premium tier costs about $29 per user per month, far above GitHub Team's $4. G…
Anthropic distinguishes workflows from agents, defining workflows as systems where LLMs are used through predefined code paths, and recommends finding the simplest solution possible, only increasing complexity when neede…
A 2025 paper from Google DeepMind found that production LLMs produce chain-of-thought explanations that contradict their actual outputs at rates up to 13.49%, with GPT-4o-mini at 13.49%, Claude Haiku 3.5 at 7.42%, and Ge…
In a blog post, developer Fabio Akita argues that 'Harness, Loop Engineering, Graph Engineering' are overhyped, claiming that as AI technology becomes commoditized, money shifts to taxonomy, creating new names for API ca…
A developer compared four Chinese AI model families—DeepSeek, Qwen, Kimi, and GLM—through Global API's unified endpoint, finding them competitive with Western models and often cheaper. The developer recommends DeepSeek V…
A new class of security threat called 'agent swarm attacks' is emerging, where multiple AI agents combine individually reasonable actions into damaging outcomes without a single malicious actor, as highlighted by a Melbo…