Anyone tried grok-bot?

Hey ZazenCodes builders, I'm Yusuf, co-founder at Vendo (YC S26). Berkeley honors grad at 19, Stanford MS, and now building open-source software that lets companies agentify their products: embedded agents your customers use to automate work, build views, and connect their tools. We're at 412 stars and I'm trying to push past 500 so more builders find it. If you have a GitHub account, this takes 10 seconds: open https://github.com/runvendo/vendo and hit the Star button in the top right. That's it. Live product if you want to poke at it: vendo.run Happy to answer anything about the stack, agents, or YC. And I'll return the favor, drop your repo below and I'll star it too. 🙏

There is this idea of self-learning in AI. Am I smarter today than I was yesterday by learning from yesterday's mistakes? Does anyone have any practical suggestions on this self-learning? I am seeing this tension. If the instruction is non-deterministic, then there are issues like prompt hacking. Besides the vulnerability aspect, there have also been numerous cases of AI agents not doing specifically what was asked. This "may or may not happen" aspect to any plain English instruction given to an AI agent. It's a request not a command. The way to solve this is to convert the instruction from plain English to some coding language and run that piece of code. Usually python works. But moving to python kills the self-improvement loop. All self-improvement options I have seen (SkillOpt for example) work on improving instructions in plain English. A request can improve over time but a command is frozen in the time it was created. Wonder how people see this tension of being more efficient today (python code) vs being smarter tomorrow (Prose in a skill)?