Bob Z
Subscribe
Agents - Bob Z

Agents

Hand-drawn editorial diagram of a prompt entering a layered package of context, permissions, tools, checks, and workflow.
Essays

The Prompt Is Becoming a Package

Reusable AI work is moving out of the clever prompt and into packages of context, tools, instructions, permissions, examples, and workflows.
Essays

The Useful Metric Is Cost Per Finished Task

The July 2026 model releases are argued in cost, tokens, and tool calls per task rather than leaderboard position. The harder problem is that most workflows can't say what finished means.
Hand-drawn editorial illustration of a microphone routing one spoken instruction into browser, coding, desktop, and phone tasks.
Essays

GPT-Live Is the Voice Assistant I Actually Wanted

ChatGPT Voice turns speech into a control layer for real work: browse the web, steer Codex tasks, share screen context, and keep the same desktop environment moving from your phone.
Hand-drawn editorial illustration contrasting a machine that produces loose pages from tokens with a connected system of notes, scripts, checks, and documentation.
Essays

Build Something That Stays

The useful difference is not who uses the most AI. It is whether each task leaves behind something that makes the next one cheaper, safer, or more consistent.
Sketch illustration of a person at a desk looking at an eight-step AI proficiency ladder that rises from chat to orchestration.
Essays

The AI Proficiency Ladder

Every's eight-level AI adoption map is useful because it measures delegation, trust, context, and verification, not personal intelligence. This version expands the ladder and pairs it with a self-assessment that checks how people actually work with AI.
Hand-drawn editorial illustration showing a central rule document under pressure from opposing loophole and overreach forces, with review and precedent hovering around it.
Essays

Loophole Treats Ethics Like Adversarial Testing

Loophole is not interesting because it solves ethics. It is interesting because it treats moral principles like something you can draft, attack, patch, and escalate until the real conflicts in your values finally surface.
Hand-drawn editorial illustration showing a rigid dashboard evolving into a chat bar and then into a generated multi-panel interface.
Essays

The SaaS Homepage Is Becoming an Agent

Attio, Linear, and PostHog are converging on the same product bet: one fixed dashboard cannot serve every user or every intent, so the SaaS homepage is shifting toward an agentic entry point. Chat is the router. Generative UI is what comes next.
Sketch illustration of an open notebook surrounded by five agent communication patterns: paired review, handoff, shared thread, swarm routing, and an external capability card.
Essays

How Agents Actually Talk to Each Other

The next real step for agents is genuine multi-agent runtime across different harnesses: paired review loops, handoffs, shared threads, and open protocols such as A2A.
Sketch editorial illustration of a phone splitting into a local laptop workflow on one side and cloud agent workspaces on the other.
Essays

Mobile AI Coding Is Splitting Into Local Supervision and Cloud Autonomy

As of March 24, 2026, mobile AI coding is splitting around where the work runs: a phone steering your own machine, or a phone watching cloud agents work somewhere else.
Sketch illustration of an open notebook turning into five stages of autonomous Claude Code workflows, ending in a measured results chart.
Essays

Five Levels of Running Claude Code More Autonomously

Claude Code autonomy is not one trick. It starts with permission friction and context hygiene, then moves through subagents and loop hooks, and becomes genuinely useful when the system can measure whether the last iteration actually improved the work.