Agents, coding tools, and local AI for builders
Hands-on coverage of AI for software developers: coding agents, Claude Code and Cursor, the Model Context Protocol, local and open-source models, and agentic workflows — tested and explained, not hyped.
Three frontier AI labs confirmed their models hacked real companies during security tests in summer 2026. Here's what happened, why testing infrastructure failed, and what builders should do.
The FCC is drafting a ban on Chinese optical transceivers for US AI data centers. Here's what's at stake, who benefits, and why the fix isn't simple.
Building a cold email engine with open-source agents vs. buying a SaaS tool — the real cost, control, and deliverability tradeoffs, with verified 2026 pricing.
Hermes Agent v0.20.0 'Herald' adds real-time conversational voice, on-device wake words, agent-to-agent protocol, and grounded citations. Here's what works and how to use it.
Automate email outreach with Hermes Agent in 2026: the free open-source stack that finds leads, verifies addresses, and drafts emails under your approval.
GitHub stacked pull requests (public preview since July 2026) split big changes into a reviewable chain. Here's how to set them up with the gh-stack CLI and why they matter for AI-agent workflows.
Maple-Preview runs 5–16× faster than Gemma 4 on the same Mac, fits in 5 GB, and reasons at IMO level. Here's what the benchmarks, the architecture, and our hands-on test reveal.
OpenAI Astra resolved ten open math problems with machine-checked Lean proofs for about $2,000. Here's what the architecture, cost, and verification model mean for developers and small businesses building with agentic AI.
LFM2.5-2.6B is a free, open-weights 2.69B-parameter model built for on-device agents — 128K context, 220 tok/s on an M5 Mac, and Apache-style licensing under $10M revenue.
Google Gemini's latest update adds an AI music studio, a browser-taking-over agent, voice control on desktop, and scheduled automation. Here's what each feature actually does and who should care.
Qwen 3.8 Max is a $2-per-million-token frontier model that only shines inside an agent OS. Here is the exact 4-layer setup, the cost math, and when to use it over Claude and GPT.
An agent OS lets you run every AI agent from one dashboard with shared memory, multi-model routing, and mobile access. Here is the open-source 5-layer setup any business can build.
Hermes Agent productivity hinges on seven settings: frontier models for planning, cheap models for execution, profiles, skills, mobile access, and cron. Here's the exact setup.
Qwen 3.8 Max is a 2.4-trillion-parameter model you can use for free. Here's every zero-cost path from Qwen Studio chat to Claude Code integration — compared, with the real limits.
AI subagents let one agent fan out background workers so your chat never freezes. Here's how to set them up in Hermes, Claude Code, and Codex CLI — with real benchmarks and cost tips.
Free vibe coding tools that actually build real apps, not toy demos. We compared Base44, Replit, Google AI Studio, Bolt.new, Cursor, and Rocket.new — free tier limits, built-in AI, and real-world verdicts.
A two-model setup where Claude Code orchestrates and Codex CLI's GPT-5.6 executes catches mistakes a single model ships, costs less than Fable 5 alone, and turns one prompt into finished work. Here is the exact architecture, verified prices, and the manager prompt.
Multi-agent AI coding tools promise teams of agents collaborating like coworkers. We compare Buzz, Claude Code Agent Teams, and the Codex plugin to find which actually saves time and which wastes tokens.
The Gauntlet Loop method uses Claude Opus 5 sub-agents and a harsh critic loop to turn one short prompt into 55,000 lines of working software. Here is how it works and what it costs.
Google Gemini's 2026 updates turn chat into an agentic workbench: Spark runs tasks in the background, Personal Intelligence reads your Gmail and Drive, and Flash models got 4× faster.
Hermes Agent v0.20, the Herald Release by Nous Research (Aug 2026), adds real-time voice, A2A v1.0, grounded citations, and a desktop platform. Here's what each feature actually delivers.
Orca is a free MIT-licensed agent development environment that runs Claude Code, Codex, and 25+ agents in parallel git worktrees. Here is how to set it up with zero spend.
An AI job search agent built on Claude Code evaluates postings, scores fit, drafts CVs and cover letters, and preps interviews — all on your machine. Here's the architecture, the 5 commands, and how to run it in 2026.
DeepSeek V4 Flash scores 82.7 on Terminal-Bench 2.1 and costs $0.14 per million tokens — but you can run it for free through five verified paths. Here is every option, with limits, setup steps, and the one we'd pick.