Build Anything With Jev, TypeSafe AI's Probability-Only Model
A screen-share walkthrough of TypeSafe AI's Jev, the token-free model that answers in milliseconds, then a live build and deploy of a Jev-powered app from idea to VPS.
September 19thCreator
A screen-share walkthrough of TypeSafe AI's Jev, the token-free model that answers in milliseconds, then a live build and deploy of a Jev-powered app from idea to VPS.
September 19thFlask's creator and Arendelle's founder on why Pi's minimal, bash-first harness is beating Claude Code and Codex, and what still has to be solved before agents work for anyone but programmers.
September 12th10x's Alex Lieberman and Dan Zakon on why the markdown you write for your coding agents is starting to matter more than the code itself.
August 20thA private, zero-open-port network for every laptop and VPS you own — set up end to end by an AI agent instead of by hand.
July 13thA two-host deep dive on why self-hosting open-source AI is a freedom fight, not just a cost play — covering hardware tiers, model benchmarks, geopolitical risk, and the case for owning your inference stack.
June 24thA senior developer's real AI-agent setup, and the argument that the harness — not the model — is where the leverage lives.
June 18thA 22-minute live walkthrough of wiring Hermes Agent to Apify MCP connectors and Supabase to automate lead scraping, scoring, and outreach.
June 15thA 32-minute live workflow session on why agentic harnesses are the right home for Fable 5, not the Claude app or raw API.
June 10thHow MiniMax M3 sparse-attention architecture makes always-on autonomous agents 10–100x cheaper than running Opus or GPT-5.
June 8thPietro Schirano left Anthropic, built MagicPath in a week, and raised funding from a single tweet. Here he explains exactly how he builds — and why he hasn't touched Claude Code in five months.
June 4thA 25-minute walkthrough of running long-lived AI coding agents on a VPS by wrapping every session in tmux — so closing a laptop, killing an SSH connection, or losing power never interrupts a job that's supposed to run for 24 hours.
May 25thA 31-minute live walkthrough of giving a Hermes AI agent autonomous phone capabilities via VAPI -- from booking a 5-cent spa call to 24/7 cold-outreach cron jobs.
May 20thSenior developer Micky Shimeles walks through the exact stack that lets him ship 95% AI-generated code without losing control: harness, context tools, service layers, and autonomous PR loops.
May 17thA 26-minute step-by-step tutorial on the agentic loop command that runs until your goal is actually done.
May 16thDavid Ondrej installs SuperGemma4-26b locally via Ollama, then open-sources a two-day Claude+Codex build: an automated loop that discovers which prompt harnesses make commercial models answer what they normally refuse.
May 11thA 47-minute walkthrough of all seven levels of Hermes Agent — from bare VPS to full MCP back end.
May 6thDavid Ondrej compresses Anthropic's official 7-hour Claude Code masterclass into 27 minutes — with two ground-truth corrections the official course is already wrong about.
January 20thDavid Ondrej installs Google's open-source coding agent, runs it live against Claude Code on a real production codebase, and lets the results speak.
June 26th 2025