The New Work Order Is Here

And it runs in the terminal

Author

Ajay Prakash
January 28, 2026

For a long time, “knowledge work” meant context switching.

Dashboards. Docs. Slack. Notion. Analytics tools. Tabs everywhere.

That model is quietly dying.

What’s replacing it is simpler, faster, and frankly more powerful:

work orchestrated through agents, with humans staying in the loop only where judgment actually matters.

Claude Code agents are the clearest signal yet that this shift is real. I’m hoping this is the last newsletter I write in the beehiiv browser and start writing these in my terminal.

To think something so cute, is about to replace me 😭

MCPs Turn Tools Into Conversations

With MCPs wired up, tools stop being destinations and start being verbs.

Instead of:

  • Logging into a metrics tool

  • Finding the right dashboard

  • Interpreting charts

  • Exporting numbers

  • Translating them into a decision

I can just ask, in my terminal:

  • “What changed week over week?”

  • “What looks off?”

  • “What would you investigate next?”

This is me just getting a sense check of our how our business is doing month on month

The terminal becomes the control plane for the business. Questions become cheap. Direction becomes fast.

This is a subtle shift, but it changes how often people actually look at data. Which changes decisions. Which compounds.

The Terminal as the New Writing Surface

Docs used to be where thoughts went to die. You’d open Notion already half-exhausted, try to sound coherent, over-edit, and end up with something vague but polished.

Now I just dump raw thinking into the terminal. Messy bullets. Half sentences. Contradictions. Directional thoughts.

The agent turns that into:

  • A clear Notion doc

  • A structured brief

  • A readable plan someone else can actually execute

Example of a document by Claude Code

The important part is not that the writing is “better”. It’s that thinking stays fluid longer.

You no longer pay a tax for clarity up front. You think first, structure later.

That alone dramatically improves communication quality.

This newsletter is not for everyone.

It’s for A+ players. People with agency, leverage, and too many tabs open.

  • 63,000 founders and operators read Ajay’s Quest.

  • 1,200+ are running companies big enough to hurt if they choose the wrong tool.

If your product belongs in that world, sponsorships are open. Reply back for details.

📈 Human-in-the-Loop Is Becoming a Design Principle

There’s a lazy version of automation and a thoughtful one. The lazy version tries to skip humans entirely. The thoughtful version is explicit about where human judgment matters.

We’ve been building workflows where humans approve intent, not execution.

Take a cinematic ad as an example.

Kudos to RoboNuggets who introduced me to this concept. LINK

The workflow looks like this:

  1. We provide a core image and character frames

  2. The agent generates scene concepts

  3. A human approves the scenes

  4. Those scenes become frame images

  5. Frames become video

  6. Video gets music

  7. Final output is reviewed before shipping

Every step is automated, but every meaningful fork has a human checkpoint.

This isn’t slower. It’s faster with taste. The system handles scale and iteration. Humans handle narrative, coherence, and “does this feel right?”

That division of labor is the real unlock.

This Is a New Work Order, Not a New Tool

The mistake is thinking this is about Claude, or agents, or MCPs. It’s about where work happens. The terminal is no longer for engineers.

It’s becoming the place where intent is expressed. Docs, dashboards, workflows, assets, decisions all downstream from that. The companies that win won’t be the ones with the most AI.

They’ll be the ones that redesign their workflows around this reality first.

⁉️ How can I get started?

  1. Get a nice terminal Download https://ghostty.org — it's fast, clean, and makes living in the terminal actually pleasant.

  2. Install Claude Code → https://claude.ai/claude-code

  3. Ask it to help you set up Seriously. Ask Claude Code how to connect your tools, configure your workflow, or write your first MCP. It'll walk you through it.

Resources to go deeper:

The trick is to stop treating it like software to configure and start treating it like a colleague to talk to.

Closing Thoughts: The Throughline

We didn’t automate work away. We compressed it.

Less friction between thought and action. So I can just speak

(Well, I hope… let’s see how 2026 goes I guess)

Until next time,

Ajay

🧠 Ajay’s Resource Bank

A few tools and collections I’ve built (or obsessively curated) over the years:

  • 100+ Mental Models
    Mental shortcuts and thinking tools I’ve refined over the past decade. These have evolved as I’ve gained experience — pruned, updated, and battle-tested.

  • 100+ Questions
    If you want better answers, ask better questions. These are the ones I keep returning to — for strategy, reflection, and unlocking stuck conversations.

  • Startup OS
    A lightweight operating system I built for running startups. I’m currently adapting it for growth teams as I scale Superpower — thinking about publishing it soon.

  • Remote Games & Activities
    Fun team-building exercises and games (many made in Canva) that actually work. Good for offsites, Zoom fatigue, or breaking the ice with distributed teams.

✅ Ajay’s “would recommend” List

These are tools and services I use personally and professionally — and recommend without hesitation:

  • Athyna – Offshore Hiring Done Right
    I personally have worked with assistants overseas and built offshore teams. Most people get this wrong by assuming you have to go the lowest cost for automated work. Try hiring high quality, strategic people for a fraction of the cost instead.

  • Superpower – It starts with a 100+ lab tests
    I joined Superpower as Head of Growth, but I originally came on to fix my health. In return, I got a full diagnostic panel, a tailored action plan, and ongoing support that finally gave me clarity after years of flying blind.