Working In Progress
This will eventually be a long form blog post.
Using AI Without Losing Your Mind
Everyone’s talking about what AI can do. Nobody’s talking about what it costs - and where the gains actually go. This is NOT a case against AI but rather practical advice from someone who is leaning heavily into using it on a regular basis.
Maybe you’ve started to kick off a few different agents. And you’re shipping faster than ever before. You’ve drank the Kool aid, and you’re a believer. But you’re feeling absolutely exhausted at the end of the day. You feel “fried”, almost like your heads been on fire. You’re eyes are tired. What the hell is going on?
The biological case for slowing down
Richard Cytowic makes this plain in Your Brain Wasn’t Built to Hold This Much Information: your brain burns ATP for every decision, every context switch, every review pass. You have a finite cognitive budget every day, and you’re spending it whether the work feels hard or not.
A Harvard Business Review study confirmed what a lot of people already felt but couldn’t name: AI-assisted work produces a specific kind of exhaustion. Not from deep focus, but from throughput. You moved fast, you produced a lot, and you’re somehow more depleted than before. What’s new? Context switching.
All about Context
No, not LLM context (which I’m sure you’re becoming an expert in). Every time you jump between tasks or check on an agent’s output, you pay an ATP toll. The pop-up saying, “I need you to make a decision” is costing you.
I’ve felt this for years before I had the vocabulary for it. Understanding how my brain handles flow and interruption isn’t just personal history, it’s what shaped how I built my setup.
As someone with ADHD I know that the pop-up, or the red circle above the app is trouble. That is, I’ve learned over years that “a quick check” is never just that. To make it worse, the fear of forgetting adds to the urgency.
The systemic trap: the vampiric effect
Steve Yegge named this in From IDEs to AI Agents as the Vampiric Effect. The cognitive overload you start to experience as you try to keep these agents funning. And one concerning aspect is that companies absorb every productivity gain you produce. The baseline just moves up. You’re not 3x more productive and compensated accordingly -you’re just expected to produce 3x now. The HBR study corroborates that. It happened with emails and smartphones. AI is just faster and the extraction is harder to see.
So is the vampire the LLM or the business who continue to push you to your physical limits? Either way, the vampire doesn’t kill you. It keeps you just alive enough to keep giving.
The Wrong Kind of Junior
Here’s a useful mental model: each agent you run is a junior engineer who wants your attention.
Now imagine you have five of them. They’re all working on something, and they all have questions, and they’re all waiting on you. Oh yea, they’re asking you for help right when you’re in the middle of something else.
That’s not a productivity multiplier. That’s a management overhead you didn’t sign up for, and it’s probably not showing up in your comp.
The people who get wrecked by this are the ones who think that AI agents are like background processes. They’re not. They require steering, review, judgment calls, and escalation handling. That’s real cognitive work. If you don’t design around it, the agents run you instead of the other way around.
What I actually do
By following my simple course for $99.99 you too can… Just kidding!
I run my agents in WezTerm which gives me the ability to split panes and use cli tools side by side. For notifications, I needed something that wouldn’t break my flow. I forked my current ZeBar theme of choice and added an “agent-deck” like feature. It adds a silent visual indicator at the top of my screen that shows agent status without interrupting me. No pop-ups, no sounds, no unwanted context switching.
The reason: notifications designed to interrupt get addressed immediately, even when that’s not what the moment calls for. That’s a panic response masquerading as productivity. The fix is simple — don’t let the tool set the terms of your attention. You check when you’re ready, not when the computer demands it.
This isn’t just useful if you have focus challenges. It’s useful for anyone trying to do deep work alongside AI agents. The goal is the same: protect the windows where you can actually get into flow.
Practical advice
Build tooling to fill your gaps, not to replace your judgment.
The goal isn’t maximum agent throughput. The goal is maximizing the time you spend on work that gets you into flow — and minimizing the interruptions that pull you out. AI should handle the mechanical, the repetitive, the first-pass. You handle the judgment, the architecture, the decisions that require context no agent has.
And watch the baseline. If your output keeps going up and nothing else does, you’re feeding the vampire. That’s worth naming — with yourself, and with whoever sets your goals.