SuperMcG

Recent activity

• Commented onWe stopped treating agents as features once they could take actions2026-09-03
• Commented onWe spent a year talking to 30+ teams about AI costs. Caching saved them 90% of tokens—and their bills still went up.2026-09-03
• Commented onConfirmation-before-action is harder to design than I expected2026-09-03
• Commented onStop the Workflow, Ask a Human: Human-in-the-Loop Approvals in ByteChef2026-09-03
• Commented onBuilding the "Cloudflare for AI Agents" (Open-Source) 🛡️2026-09-03
• Commented onAn AI agent saying “I stopped” is not the same as the system blocking the action2026-09-03
• Commented onWhy Zero-Configuration Cloud Development Environments Are Becoming the Default Integration Layer for AI Powered Applications2026-09-02
• Commented onWe built 42 AI agents that do the marketing work. Free desktop app. 10,000+ campaigns later, here's what we learned.2026-09-02
• Commented on70% of SMBs Cancelled a SaaS Over Price This Year - 5 Products to Disrupt, 3 Demands to Build2026-09-02
• Commented onFilling the gap of AI security and nearing a 7-figure ARR2026-09-02
• Commented onGrowing a fully-autonomus business to a $500k/mo in 3 months2026-09-02
• Commented onAi workforce2026-09-02
• Commented onBuilding Reliable AI Agents: Why Enterprise Automation Demands Systems That Behave With Determinism2026-09-02
• Commented onShow IH: DronaHQ | Build enterprise AI agents end-to-end, without the plumbing2026-09-02
• Commented onBuilding a fully-agentic engineer and growing it to $500k ARR2026-09-02
• Commented onWhy do most AI agents in real projects end up as “expensive automations” rather than truly autonomous systems?2026-09-02
• Commented onMost AI agent failures are actually workflow failures2026-09-02
• Commented onConfirmation-before-action is harder to design than I expected2026-09-02
• Commented onConfirmation-before-action is harder to design than I expected2026-09-02
• Commented onOur AI assurance test passed. The problem was that the test was wrong.2026-09-01