I've decided to take back control of my brain in the era of ai

Software engineer Zach, who has used AI agents daily for over a year, warns that excessive delegation of coding and writing to AI is eroding developers' own thinking and understanding. He describes a 'very specific brand of laziness' and 'agent FOMO' that leads to a loss of deep knowledge, and announces a personal project to reclaim his cognitive abilities by writing and building without AI assistance.

to write is to think. think for me. whir-buzz-and-spin until the content is complete. while I stare blankly at the terminal fleet. it's ok, i say. another task done fast. i'll use my brain next time. but for now, this will pass. write a quick script, run, remove the em—dash. actually, too lazy for that. claude, pls remove them /fast. post the content everywhere, linkedin, reddit, i don't care. back to agent number 42, wait, what am i working on there? Hey, I'm Zach. I'm a software engineer not a poet in the contact center software space, and above is what the current state-of-things feels to me everywhere I look on the internet. AI generated content. Em dashes galore. Delegation of entire plans, features, and important system design to agents on x-high effort. I've been writing code for over 7 years, and over the last year or so I have used AI to plan and implement code every day. Like many of you, I started out small, using AI to rubber duck and validate ideas. I gradually moved towards letting AI agents write more and more of my code base, tests, deployment scripts, and even the planning of large features. Over time, I started to realize just how much I was relying on AI for day-to-day work. From coding, to writing content, to designing systems. Reaching for ways to build diagrams with AI, generate content I could use on posts, the list of "things I could save time on using AI" grew bigger and bigger. Looking back, I remember and cherish the amazing things I was able to accomplish in implementations and projects without having access to AI, all while feeling like I had a complete understanding of the machine underneath. As I'm writing this today, I recognize and would like to call out a huge problem that myself and many other developers are starting to realize: a very specific brand of laziness and delegation of our own thinking and writing ability that comes with the excessive use of AI https://www.youtube.com/watch?v=iqk2SL4fmNk&t=53s . I'll just use it for research, then I'll design the feature my self... Well, the research is done, but I probably get this feature done a lot faster if I just use an agent loop. Well, I guess we could also have it write all the migrations and help with the deploy pipelines for it. A few days pass, your infra / tech lead are on a call with you, asking questions about the design So, why did you decide to build it this way? Did you think about the fact that our Service ABCalready handles this part? Meanwhile, you have 3 agents running, planning and implementing unrelated work, and mid-conversation you suddenly get agent FOMO and context switch to make sure there aren't any pending implementation decisions in your agent loops. You freeze up. You want to collaborate, but you CANNOT recall how this thing was built, and on top of this, your 3 agents' tasks are taking up your headspace that you should be using for the current discussion. You open the plan file and try to fumble your way through exactly how it was implemented. Where was the mermaid diagram that came with this feature? Is it up to date or did it drift during the implementation based on an agents finding during the implementation? Sound familiar? This scenario demonstrates 1 the cost of delegating writing and thought to the agent and 2 having a constant sense of "agent FOMO" - are we running as many agents as we can make the most of our plan usage as possible? Our attention is limited. Our brains have a process of learning that involves writing, hands-on involvement, and trial and error that we're actively tossing out the window. Writing is a form of thinking. If we delegate our writing, we delegate our thinking. I've decided I am not going to let that happen. So what am I doing about it? I've decided to start a project to take back control of my brain in the era of ai . Yeah, the content will not be perfect. It will be human. I will make mistakes. Some of my thoughts may not be perfectly crafted. But it will be mine, in my own original voice. It's time for us to reclaim our brains, and think and write again in an original way. Interested in joining me?