One of the most followed software educators just dropped a podcast. I watched it. STEAL these INSIGHTS for Claude Code.

So I just watched one of the most followed software educators on the internet just dropped a podcast. I watched it. STEAL these INSIGHTS for Claude Code. And his take on AI is genuinely different to what everyone else is saying. Okay, here are the top five insights that are worth stealing. The first one is, everyone is focused on the wrong thing. Everyone is focused on the model, or benchmark it beats, what model is it better than. But his take on it is that the model is only half the equation. The other half is your harness, your setup, your Claude and D file, your environment. You almost have no control over the model, but you do have control over the environment it operates in. So you should focus there. Second insight is that AI has completely eaten the tactical work. You know, writing the code, fixing the code, debugging, all of that is gone. AI can now do it better and cheaper. Your value is now entirely in strategic thinking. Decide what gets built and how it gets built. If you're still in the weeds of things and doing the tactical work, you're basically trying to do the AI's job. Insight. The third insight is, stop bloating your setup. His advice was just delete everything. Delete your Claude and D, remove every MCP, plugin, etcetera. Start from a clean setup and watch what the agent does naturally. Most people have too many instructions, connected tools, that the agent gets confused before it's even started. And then only add something back when you genuinely miss it and you cannot work without it. No. And number four, think in queues not loops. This links into the strategic thinking. So for example, you give the agent a task, it completes the work, you review it, then you decide, okay, where to go from here. So it's not an everlasting loop, it's more of a queue. So you still have control over the strategic direction in which the agent takes. That's great work. All great teams have always worked. And number five, your skills are your ceiling. Model cannot go above what you know. Okay, which then links into AI amplifies your current skill set. It can't go above your skill set. It always matches it, but just does it at a faster rate. So if your understanding of what good is, is limited, then the AI cannot compensate for that. So the people who are getting the most out of AI are the ones who are best at their own craft, not the ones with the best model. So if you do want to break into a new field, I've got a video on how you can do that. It's always about the information, whether it's through you or through the AI first, building the video. Okay, because a video editor is always going to beat editing with Claude. But those are all the insights. Hopefully, they're insightful. Use them, steal them, let me know how it goes. Follow for more free game.