Full-text search across 61880 articles. Combine with topic and date filters; results sorted by relevance.
Snyk's new VulnBump benchmark found that on the same Claude Sonnet 5.5 model, Snyk's Evo Remediation Agent passed 90.2% of 307 real security-driven dependency upgrades, versus 77.5% for Anthropic's Claude Code with Snyk'…
AI is flooding GitHub with low-quality repositories, making it harder to trust open-source code, according to a Czech developer's analysis. The ease of generating thousands of lines of code via AI has led to 'project inf…
Cursor published CursorBench 3.1, a benchmark for AI coding agents, revealing Anthropic's Fable 5 models as top performers and Cursor's Composer 2.5 achieving near-top performance at 3% of the cost. GitHub made Kimi K2.7…
Google DeepMind released EmbeddingGemma 2, a 740-million-parameter open embedding model that maps text, code, images, video, and audio into a single 768-dimensional vector space and runs in 567MB of RAM quantized on-devi…
TAUSIK, a verification layer for AI coding agents from maintainer Kibertum, released version v1.11.4 on Oct 12, 2026, gating task start and completion on specifications and evidence via MCP tools, skills, and host-depend…
A study out of MSR '26 examining 3,858 pull requests found that reviewers responded more warmly to agent-written code than to code written by people, according to an account of the research. The author argues that adding…
Two engineering teams reported that AI coding agents shifted their bottleneck from code generation to code validation. Linear reworked its CI pipeline after test suites nearly quadrupled, cutting pull request wait time f…
Bend, a high-level programming language that compiles mathematically verified code to run natively and in parallel on both CPUs and GPUs without manual multi-threading, is pitched as a constraint layer for LLM agents gen…
A developer maintaining the open-source decision-history tool Selvedge outlined a Markdown-based workflow for recording why a coding approach was rejected and under what evidence it should be revisited. The method stores…
An Anthropic engineer stated that production code written by Claude should meet a higher bar than human-written code, citing lint rules, tests, Claude-driven end-to-end tests, Claude-powered fuzzers running daily, automa…
A developer argues that AI-generated code is invalidating traditional software quality assumptions, warning that measuring bug rates early in the AI coding era may be misleading. Because AI is likely to produce far more …
Microsoft's Engineering Thrive initiative, led by Corporate Vice President Tim Bozarth, finds that AI is shifting engineering bottlenecks from code creation to planning and validation, with verification and confidence be…
AI coding agents are shifting the target of abstraction from setup and repetition to the act of thinking through solutions, raising the question of whether developers will lose the need to code. The author, drawing on th…
A developer has released frontend-standard-skills, an open-source collection of installable AI skills for Claude Code, Cursor, and other AI coding tools that enforce consistent, accessible, and performant frontend code a…
PrestaShop added a hierarchical `.ai/` folder to its core repository, creating an AI agent onboarding system versioned directly alongside the code. The system includes 57 domain contexts, 22 component contexts, pre-compu…
The article argues that the primary challenge for AI coding tools shifts from code generation quality to context management as projects scale beyond 40 files. Key differentiators include repo understanding, multi-file co…
A LangChain4j experiment directed an LLM-backed code assistant to use the framework's documentation to design and implement a multi-agent coding system that could write, test, and debug code. The self-built agent fixed r…
Boris Cherny describes a growing trend of developers building harness-level loops on top of coding agents, where tasks persist beyond a single agent session. He expresses concern that this approach amplifies models' tend…
Zhipu AI's GLM 5.2 coding model can be used as a cost-effective backend for AI agents via OpenRouter, offering strong code generation and reasoning at 10-20% of the cost of Claude Sonnet or GPT-4o. This guide provides st…
The article explains how to use the `Microsoft.Extensions.Telemetry.Abstractions` source generator to simplify metric creation in .NET, replacing manual boilerplate code from the `System.Diagnostics.Metrics` APIs introdu…