AI — Page 32

Kagenti's Identity-Based Security for Multi-Agent AI

Kagenti tackles the confused deputy vulnerability in multi-agent AI with SPIFFE, AuthBridge, and chain-aware delegation. Here's how it actually works.

Brick-textured "CLAUDE" and "CODE" letters on dark background with yellow "SubAgents 2.0" banner and red "NEW" label

Claude Code Nested Subagents: Power and Cost Explained

Claude Code Nested Subagents: Power and Cost Explained

Anthropic's nested subagents let Claude Code spawn agents five levels deep. Here's what that actually means for your workflow—and your token bill.

Person wearing glasses against dark background with purple code diagram and "think series" branding, discussing AI pair…

AI Pair Programming: Productivity Tool or Security Risk?

AI. Rachel "Rach" Kovacs4 months ago· 6 min read

AI Pair Programming: Productivity Tool or Security Risk?

AI pair programming promises faster code and fewer bugs. But what happens when your AI collaborator is confidently wrong about security? A practical read for developers.

Three men discuss Markdown against a vibrant purple background with yellow text reading "MARKDOWN IS **EVERYWHERE**" and…

How Markdown Went From One Guy's Blog Tool to AI's Native Language

How Markdown Went From One Guy's Blog Tool to AI's Native Language

John Gruber invented Markdown in 2004 to make blogging less painful. Two decades later, it's the formatting language every LLM defaults to. Here's how that happened.

Developer at neon-lit desk with GitHub homepage displayed on dual monitors, surrounded by programming code and tech neon…

35 Open-Source Tools Shaping AI Dev in 2025

35 Open-Source Tools Shaping AI Dev in 2025

GitHub's latest trending repos show developers wrestling with token costs, agent reliability, and AI tooling fragmentation—here's what's actually worth your attention.

Diffusion Gemma Runs Locally—and That Changes Privacy

AI. Rachel "Rach" Kovacs4 months ago· 7 min read

Diffusion Gemma Runs Locally—and That Changes Privacy

Google's Diffusion Gemma runs on consumer GPUs at 700+ tokens/sec. For privacy, the real story isn't speed—it's that your prompts never leave your machine.

Developer at desk viewing Hacker News website with code editor, surrounded by neon green Y Combinator branding and glowing…

Hacker News Digest: June 12, 2026

AI. Rachel "Rach" Kovacs4 months ago· 8 min read

Hacker News Digest: June 12, 2026

From a $6K AI AWS bill to Meta's facial recognition playbook, Hacker News surfaced the tensions defining tech in June 2026. Here's what mattered.

Man in dark polo shirt smiling at camera with neon design sketches on black background and "Design To Code with MCP" text…

How MCP and AI Agents Are Reshaping Software Design

How MCP and AI Agents Are Reshaping Software Design

IBM's Will Scott explains how design systems, context engineering, and MCP are combining to let AI agents build software that actually follows the rules.

DiffusionGemma Generates Text Like an Image Model

AI. Yuki Okonkwo4 months ago· 7 min read

DiffusionGemma Generates Text Like an Image Model

Google DeepMind's DiffusionGemma borrows from image diffusion to generate 700–1,000+ tokens/sec. Here's how the architecture works—and where it falls short.

White text reading "loop engineering" centered on a black background with a vibrant purple-to-cyan gradient border

Loop Engineering: Moving Beyond One-Shot AI Prompting

Loop Engineering: Moving Beyond One-Shot AI Prompting

From cron-job automations to multi-day autonomous goals, loop engineering is changing how developers interact with AI. Here's what that actually means.

NVIDIA Mini Datacenter presentation with person standing before compact server unit highlighted with green accents and key…

Span's XFRA Node Wants to Put a Data Center in Your Yard

Span's XFRA Node Wants to Put a Data Center in Your Yard

Span and Nvidia want to bolt $250K of AI computing hardware to the outside of homes. The pitch is clever. The fine print is worth reading carefully.

Aerial view of a massive data center facility in Utah with rows of blue server units on a white roof, surrounded by…

Rural America's Revolt Against AI Data Centers

Samira Barnes·4 months ago

Rural America's Revolt Against AI Data Centers

Bipartisan opposition to AI data centers is reshaping local politics across America. From Utah to Georgia, communities are fighting back over water, power, and broken promises.

ChatGPT and Codex logos connected by an arrow against a purple gradient background with "FINALLY FREE" text at top

OpenAI Codex Gets Banked Resets for Developers

AI. Samira Barnes4 months ago· 7 min read

OpenAI Codex Gets Banked Resets for Developers

OpenAI's banked resets update for Codex lets developers save usage allowances for when they need them. Here's what the change actually does—and what it signals.

Woman in floral dress gestures while presenting beside abstract black circle artwork in wood-paneled studio setting

Anthropic Launches Claude Fable 5 at Tokyo Keynote

AI. Yuki Okonkwo4 months ago· 7 min read

Anthropic Launches Claude Fable 5 at Tokyo Keynote

Anthropic unveiled Claude Fable 5 and Mythos 5 at its Tokyo keynote, with new managed agents, dynamic workflows, and a novel approach to AI safety guardrails.

Leadpages Auto-Generates Brand Kits from Landing Pages

Alex Volkov·4 months ago

AI. Alex Volkov4 months ago· 6 min read

Leadpages Auto-Generates Brand Kits from Landing Pages

Leadpages can now extract a full brand kit—colors, fonts, imagery, and tone of voice—directly from an existing landing page in about a minute.

Man in glasses next to file folder diagram with loading icon and connected pixel characters labeled "Subagents 3.0?

Claude Code Now Supports Nested Subagents

AI. Rachel "Rach" Kovacs4 months ago· 7 min read

Claude Code Now Supports Nested Subagents

Anthropic added nested subagent support to Claude Code. If you're already using subagents in your workflows, here's what changes—and what new risks come with it.

Large "LIFE OS" text with arrow pointing to four blue-outlined boxes listing Memory, Skills, Workflows, and Goals against a…

PAI Gives Claude Code Persistent Memory and Structure

AI. Yuki Okonkwo4 months ago· 7 min read

PAI Gives Claude Code Persistent Memory and Structure

PAI adds persistent memory, custom skills, and structured workflows to Claude Code. Here's what it does well, what it costs you, and who actually needs it.

Man in blue shirt discusses AI model optimization with overlaid graphics showing rLLM-FinQA-4B training framework and Qwen3…

A 4B Model Beat a 235B Model for Under $500