Full-text search across 58891 articles. Combine with topic and date filters; results sorted by relevance.
A pre-registered, five-model paired evaluation tested a specification frame for LLM-generated code on money arithmetic, time handling, retry safety, and access control tasks, according to the study's authors. The evaluat…
Jevia launched as an outcome-based model router for coding agents that routes a task to a model tier and uses verified outcomes to improve future decisions. The tool installs as a prebuilt binary without Rust or Cargo, a…
Researchers introduced WideSWE, a benchmark for evaluating coding agents on coordinated changes across multiple repositories, arguing that existing coding-agent evaluation remains largely confined to a single codebase. T…
Ali and Vrajang released Tenjin, an alpha-stage tool router for Claude Code built on Jev that watches masked prompts, session context, web searches and subagent tasks to supply tools from a curated library including Exa,…
CrossCheck launched a free Quick Scan tool that inspects a public GitHub repository's root package.json and six known lockfile paths to flag conflicting package-manager signals, which the company says can give coding age…
Coral Bricks, a Seattle-based inference platform founded on 2026-06-12 and backed by Afore Capital and Foundations Accelerator, launched with a claim of 469 tokens per second for DeepSeek Flash v4.1 on coding workloads. …
A software developer built a macOS Swift app that captures cursor position, on-screen pointing gestures, and synchronized voice input to generate context briefs for AI coding agents like Claude Code and Cursor. The tool …
Metermaid AI scored 341,054 public coding-agent trajectories from 11 public datasets, including SWE-smith, SWE-rebench, Open-SWE-Traces, mini-coder-trajs and agentic-coding-trajectories, for loops, blind retries, oversiz…
Hacker News user Marvin_RunAI posted an Ask HN thread titled "How many runs before you trust a coding agent's result?" that had drawn 1 point and 1 comment roughly 37 minutes after posting. The question asks the Hacker N…
DSCI, a self-hosted Git and CI server, now offers an LLM-based pipeline code generator that produces CI configurations from natural-language requests, according to a post on Hacker News. Users submit a prompt such as "Bu…
Codacy is dogfooding its Verity AI agents to address dependency version selection risks in AI coding assistants, according to a 2026 report on the practice. The report notes most teams working with AI coding agents still…
PINNStudio, a free, open-source no-code graphical user interface for setting up, training, and visualizing physics-informed neural networks (PINNs), has been released to help researchers with limited coding experience. T…
A user reports using Fable 5.1, an AI coding tool, to fix the BTRFS RAID56 write hole after experiencing data corruption twice, claiming the tool passed state machine tests but seeking community review due to low coding …
Meta AI released Muse Spark 1.3, an agentic coding model that uses approximately 20% fewer tool calls and 25% fewer tokens than its predecessor Muse Spark 1.2. The model is designed to improve efficiency in coding tasks …
A Hacker News commenter argues that vibe coding the frontend is acceptable for a DevOps expert, provided they can evaluate UX and accessibility, since LLMs excel at frontend code and the frontend can be decoupled from ba…
Cursor launched Router on July 22, 2026, a request-level classifier trained on over 600,000 live coding sessions that automatically selects the AI model for each prompt, cutting AI coding costs by up to 68%. In productio…
A developer claims that routing Claude Code requests to alternative models such as DeepSeek or Grok can cut costs by 60%, based on a blog post from Leanroute.dev. The post describes using the Leanroute.dev MCP to impleme…
A developer has launched IndiePaste, a fast, ephemeral code sharing tool built with Next.js and Upstash Redis. The tool offers burn-after-read controls, automated expiration, and integrated Groq AI for one-click code ana…
Code Stitcher, a new tool launched on Hacker News, lets developers apply any LLM-generated code output directly to their local codebase, supporting Python AST checking and all GDScript (Godot) files. The open-source proj…
Ast-grep, an open-source structural code search tool, now lets developers run ast-grep pattern syntax searches across public GitHub repositories, enabling pattern-based code search beyond simple text matching.