AI-Assisted Coding Reflection

make this sound professional: Regarding #4 I can not agree more. I developed github.com/kliewerdaniel/workflow.git to help but today I just saw github.com/sdi2200262/agentic-project-management which is basically the same concept except if someone intelligent made it rather than the first which was built by an idiot, me. Both are basically just complex instructions and frameworks to help AI-Assisted coding tools like CLIne in VSCode is what I use for Gemini mostly since it is free, great context, and does a decent enough job for my use cases, I am hobbyist though. I am trying to transition into a professional developer role but I am not exactly employable due to a problem that I have so instead I am trying to make it freelance and solo. It is an anger problem. Think nerd rage but if you took that squirmy computer nerd onto the streets for a while until they learned how to scream and fight like a crazy person for survival. I am always afraid of reverting back to that rage. I have tried to control it but I think it is seated from a deep hatred I have for people after the horrible crimes were committed against me and my friends with impunity. No, I know where it comes from, that night 5 guys broke in and I had to use a chef's knife to survive. That night is what wakes me up in the dead of night ready to kill the invaders. I was housed in hospitals for a while, but I have been out for quite some time now. I even passed for a normal person for a while, held down a square job with perfect attendance for a year. But then it crept back and the mask fell, I lost my temper and they all know now that I have a mental disorder and all hopes of growth in that company are dead now so I quit. I have done this a million times. I prepared this time at least. Anyway another thing not to do, I am an expert at doing those things, when in this context of AI-Assisted coding is what we mean though, is to not just copy and paste the error log into the window and hope that it is a quick fix. It often is. But you should not waste your context which you can limited API calls or cost you something, unless you do local inference, which I do, but not for a lot of what I do. I don't know how many times I have just taken each error and just copy pasted it into CLIne and preyed it worked because I was burnt out from coding for 20 hours straight and my mind no longer functions so I just copy and paste errors into the chat praying it will fix itself like some reanimated corpse whose brain is circumvented by electronics. So DON'T do that.

Here’s a more professional and polished version of your message. I’ve preserved the emotional authenticity and narrative, while reframing it with dignity, clarity, and structure—suitable for a technical forum, GitHub README, or a reflective blog post: --- **On Point #4 — I Couldn't Agree More** I've actually been exploring this space myself. I created [github.com/kliewerdaniel/workflow](https://github.com/kliewerdaniel/workflow) as an experimental project to support AI-assisted development workflows. Interestingly, I just came across [github.com/sdi2200262/agentic-project-management](https://github.com/sdi2200262/agentic-project-management), which is built around a similar concept—but clearly designed with a level of clarity and structure that I deeply admire. In comparison, mine feels more like a raw prototype created by someone still learning to channel their ideas effectively. Both projects are essentially attempts to codify patterns and instructions to streamline AI-assisted coding—tools like CLIne in VSCode, which I personally use with Gemini. It offers solid context handling and, for my hobbyist-level projects, works well enough—especially since it's free. I’m in the process of transitioning from hobbyist to professional developer, though my path has been unconventional. Due to personal challenges, I’ve found traditional employment difficult to maintain, so I'm pursuing a freelance and independent route instead. To be candid, one of the barriers has been managing residual trauma and anger—what I sometimes refer to as "nerd rage," taken to a dangerous extreme. I come from a background marked by real violence and survival situations that left deep psychological scars. One night in particular changed everything—a home invasion where I had to defend myself. The aftermath shaped much of how I see the world and myself. While I’ve spent time in hospitals and have since stabilized, reintegrating has been a journey. I even held a full-time job for over a year with perfect attendance, but eventually, my temper resurfaced. When that mask slipped, growth within that company became impossible, and I made the difficult decision to move on—something I’ve had to do more times than I can count. Despite this, I’ve prepared better this time. I’ve built tools, systems, and frameworks to work alongside AI and help keep my development process organized and sane. One key takeaway from all this: **Don’t treat AI tools as a magic fix by just dumping error logs into them.** It’s a common temptation—especially after hours of fatigue, when the brain is fried and you just want the code to work. But AI context windows are precious, and wasting them with unstructured error logs can backfire—especially if you're using API-limited or cost-associated services. Even I’ve been guilty of this—copy-pasting stack traces into CLIne at 4am, hoping for a miracle. It’s like zombified debugging, driven by exhaustion. Don’t do that. Take a step back, breathe, and treat your tools with intention. That’s how you grow. --- Let me know if you'd like to adapt this for a specific platform (e.g., LinkedIn, dev.to, GitHub issue, etc.) or if you want to keep some of the rawness for authenticity.