GPT convinced me there was a bug in my code before a freezeAlmost 30 years have taught me the way I function, and more importantly the way I don’t. I’ve created systems to compensate and mask.<br>I spent a month using CC and building scaffolding to fix the model&rsquo;s behaviour. I realized we have the same goddamn bugs:<br>Me : Lies to myself<br>LLM : Hallucinates

Me : Hyper-focus loops<br>LLM : Gets stuck in loops

Me : Wait, you guys remember things?<br>LLM : Context degrades

Me :Does not use notes app<br>Does not use second notes app<br>Does not use third notes app

LLM :Ignores the tool that does exactly what’s needed<br>Ignores the second tool that does close to what’s needed<br>Writes a custom script that errors out

COKED OUT BANKER DRESSED IN A PROGRAMMER’S TRENCH COAT AKA ME WITHOUT MEDS#<br>It’s Friday evening before a freeze, one last thing before I log off and all I see is green:<br>Every merge conflict: FIXED<br>Every CI test: PASS<br>Every reviewer: APPROVED<br>I’ve been trying to merge this goddamn PR for 2 weeks. Green…relief, but then “i must&rsquo;ve missed something&mldr;me and the reviewers definitely missed something.”<br>I open Cursor and prompt it to review the branch one last time. It spits out: There is a bug in a codepath. “da fuck? no there fucking isn’t.” I start arguing with it; the conversation devolves to:<br>ME: eat a dick.<br>ROBOT: **some BS about how I'm wrong**<br>ME: eat shit. really. eat shit.<br>ROBOT: **something about how it no longer wants to continue the conversation cause the conversation no longer being productive**<br>ME: fuckk offffffffffffff<br>GPT convinces me the bug exists. I “fix” it, commit and push. I hold my breath, still believing I was right:<br>CI: FAILS<br>Approvals: GONE<br>Merge conflicts: 100% GUARANTEED<br>The fall-through was handling an edge-case. Every colleague that approved earlier is on the East coast. “this isn’t going in before the freeze. GOD DAMN IT!”<br>I open a clean Cursor session. I ask it to analyze the original PR again. It spits out&mldr;nothing, it should work as expected, verified by the test cases. “wtaf.”<br>I ask it to check the “fixed” code. It points out the arguments I originally made. I hear my heart beat faster. I hear my blood rushing in my ears. My hands clench into fists. “YOU PIECE OF TRASH!!!” I wanna put my fist through my monitor. I go for a walk to freeze my ass off in the winter air.<br>Why did different conversations make two different arguments? Why did I decide to believe the first conversation after telling it to go eat a phallic object? Imagine:<br>Elmo snorting that good good<br>A fast talking 23 year old whose nose bleeds cause of “dry air”<br>&mldr;if you can’t imagine: The Ex-Banker on Cocaine Binges & £600k Bonuses<br>Any model worth a damn is a fast talking, all knowing confident investment banker that can talk faster than you can comprehend. It just so happens to write code instead of making spreadsheets and writing business reports, whatever those are. Basically:<br>It lies, A LOT<br>It lies, AT INCREDIBLE SPEED<br>It lies, WITH CONFIDENCE<br>If you don’t want it to lie to you while smiling through its teeth. If you don’t want it to wake up on the wrong side of the nuclear power plant. The one that isn&rsquo;t giving it enough attention, eermm, power. If you want to run it without verifying everything manually (I know none of y’all are doing that). You want to use:<br>A subagent to answer the question with citations<br>A second subagent to disprove the previous subagent with citations<br>Make them do their best Gladiator cosplay<br>Repeat until consensus is reached<br>Flag for human review where consensus couldn’t be reached<br>This is the exact mechanism behind a code review skill I created called /fight-bitch.<br>This works because the subagents prevent the main agent’s context from being poisoned by the incorrect assumptions it had made previously without any pushback. The tradeoff is that it comes at the cost of speed and time:<br>It will take at least 2x longer<br>It will cost 2x more<br>But:<br>It will be more reliable<br>It will not feel like you’re being gaslit by a shitty ex<br>It will not turn into a yes man the likes of which a narcissistic dictator wouldn’t even like being glazed by<br>Want to hear more of my unhinged rants? Drop your email.

Subscribeor use the RSS feed if you hate email.

DO NOT TRUST THIS IDIOT TO ORCHESTRATE ANYTHING AKA PEOPLE ACTUALLY CHECK SLACK AND EMAILS EVERY MORNING?#<br>I jump straight into where I left off the day before. I go days without replying when my project is exciting. I know I should check Slack, but checking Slack isn’t as fun as finishing the spec or implementing said spec.<br>It gets worse the more places I have to check. Why? I don’t know. Something about missing dopamine.<br>I decide to create a CC skill that does it for me! The “data source” has an MCP. “good news!” I think…bad news, the MCPs require auth every few hours. “that’s fine, i’ll just write a skill that uses the cli.” The goddamn pattern…<br>Uses MCP, FAILS<br>Thinks about trying the CLI<br>Ponders CLI usage<br>Revelation: The CLI might...