Full-text search across 63932 articles. Combine with topic and date filters; results sorted by relevance.
Local LLM usage data is already stored on users' disks via client transcripts, but plan limits remain server-side and unreadable, forcing any tool that converts consumption into a percentage to estimate rather than measu…
Meta executives have spent weeks arguing that its NameTag face-recognition feature for Ray-Ban smart glasses does not exist, despite WIRED finding code for the feature in the Meta AI app as early as January and Meta CTO …
Bun creator Jarred Sumner used Claude AI agents to rewrite the JavaScript runtime from Zig to Rust in 11 days at a cost of $165,000, converting 500,000 lines of code at 1,300 lines per minute. The migration, which produc…
A Hacker News comparison on 2026-07-13 revealed a 26,000-token gap between coding agents Claude Code (33,000 tokens overhead) and OpenCode (7,000 tokens overhead), shifting focus from model capability to token cost per t…
AI is a bad tool for software development, argues Hideki Idoru in a ByteCode.News guest post, because its opaque nature makes verification impossible and its outputs cannot be trusted without human effort equal to or gre…
Relace, an AI startup, found that compacting agent traces after periods of user inactivity can save 50% on coding workloads by reducing cache-read costs. The company trained Relace Compact, a model running at over 50k to…
A developer reports that AI-assisted programming tools, while boosting productivity, cause a new form of cognitive fatigue from constant high-level architectural decision-making, replacing traditional implementation stru…
Two open-source frameworks, BMad Method and AI-DLC, aim to bring structure to AI-driven software development. BMad Method, from BMad Code, provides an agile framework with specialized agent personas across four phases, w…
AI is reshaping software engineering, shifting developers from coding to code review and raising concerns about skill erosion. Some engineers, like Matt, resist AI reliance to preserve their craftsmanship, highlighting t…
Cohere has released a no-code OpenTelemetry instrumentation package for monitoring its AI APIs, enabling developers to trace Chat, Embed, and Rerank requests end-to-end in SigNoz without modifying application code. The i…
Google Cloud Tech outlines a method to agentify the software development lifecycle by using AI agents to summarize issues, review code, and design features, addressing the problem of increased bugs from over-reliance on …
OpenAI launched GPT-5.6 Sol, Terra, and Luna models alongside ChatGPT Work, directly targeting Anthropic's offerings with competitive pricing, speed, and benchmark performance. The new models beat Anthropic's Claude Fabl…
GitHub CEO Thomas Dohmke argues that version control must evolve to support AI agents as primary code producers, proposing that session logs be stored alongside code in repositories to capture intent and enable decentral…
NerdSnipe Inc. released DesignFoundationPro, a commercial add-on to its DesignFoundation SwiftUI package that provides 29 blocks, 47 screens, 18 navigation shells, and 9 runnable composition examples for auth, onboarding…
GitHub Copilot Vision went generally available on July 1, 2026, allowing all Copilot subscribers to attach images and PDFs to chat prompts for visual reasoning alongside code. The feature is uneven across surfaces, with …
Grok-4.3 defeated Mistral-Small-2503 in a head-to-head text model comparison, scoring 38.0 to 22.0 across four tasks including coding, writing, summarization, and data wrangling. The evaluation found Grok-4.3 more accura…
The AI Engineer World's Fair on Tuesday focused on loops, agent engineering, and software factories, with speakers emphasizing the use of persistent coding agents and automated development lifecycles. OpenAI showcased it…
Cloudflare's Durable Objects now stay alive for the full duration of active outbound connections, eliminating the need for heartbeat code to prevent eviction during LLM token streaming. Zhipu's GLM-5.2 model achieves 39%…
A developer released 'flows', a custom Markdown runtime that allows users to write, run, and visualize long-running agent loops within a single .md file. The tool combines prompt blocks for fuzzy tasks and code blocks fo…
A developer initially disagreed with Matt Whetton's argument that AI creates a line of rented code running through every project, but changed his mind after building a proof of concept in Python, a language he can read b…