I am looking for a way to run different chat sessions on different branches. Example working on docs, database and scripts at the same time without having a cultured source control window. Has anyone found a solution to that? I had the same issue with CLI tools.

Does anyone have any solution to this malware risk of using Cursor? This is the main reason I am mostly using VScode and Gemini CLI right now. I'd use cursor more if I had a solution to this.

Hello folks, The other day i was thinking about AI app use cases for majority of businesses that we study here and how to use the course to build an app. So what i could see is that majority of the use cases fall into 3 categories Chat / Chat+RAG and AI Agents for x and y. Correct me if i am wrong there may be edge cases ( mine is i think) . I would like to know your opinions and if this bucket logic works or not. Also since I think 🤔 my use case is an edge case i cannot wrap my head around the vision problem. The thought comes from developing a solution for my own domain of construction as we need to do read pdf drawings which are plotted from cad applications. Also the data is available in excel sheets. The thing is i am not sure about the selection of ai model selection for vision purpose since i tested this with gemini chatgpt or claude and all of it majorly failed to extract the data against my instruction prompt so the whole logic of app fails. In such cases how can i approach this problem? Any thoughts ideas? Fine tuning etc? Also system architecture?

Claude-4 is incredible! I’ll keep recommending it as the #1 model because the reasoning and depth it brings to coding projects is hard to beat. The challenge is that I’m building 9 projects at once right now. When you are coding daily at that scale, even the best models start to add up in cost. So I went hunting for a lighter option I could lean on in Cursor. I tried Gemini-2.5-Pro. I tried GPT-5. Both solid. Then I tested grok-code-fast-1. Here is what I found: - Way faster than Grok-4 (they were not kidding with “fast” in the name) - Free right now in Cursor - For raw speed and responsiveness, I have been enjoying it more than Gemini-2.5-Pro and GPT-5 If you are coding on a budget or just tired of waiting around for long completions, this is worth trying. I will keep sharing how it holds up as I push it across my projects. If you try out the model, I'd love to hear what you think about it!

Gemini CLI just dropped 7 major updates and if you're exploring AI dev tools, this open-source project is worth a look. In this new video, I break down all 7 new features inside Gemini CLI 2.0 — from editor integration to GitHub PR reviews — and share what’s actually working and what still feels unfinished. 🧠 What you’ll learn ✅ Gemini CLI Companion — connect Gemini directly to Cursor or VS Code ✅ Init Command — auto-generate a full project memory file with one prompt ✅ Persistent Rules — train Gemini to follow your coding preferences ✅ Stats Mode — track model usage and tool calls per project ✅ Multi-Workspace Mode — pull from multiple codebases at once ✅ Chat History — save, resume, and name CLI conversations ✅ GitHub Actions — PR review automation (still a bit buggy, but promising) Get the latest version of Gemini CLI here: 👉 brandonhancock.io/ai-driven-development P.S. Want to take AI-driven development even further? 👉 Pre-order here: shipkit.ai What is Shipkit? • Prebuilt templates for the most common AI apps: Chat, RAG and ADK agents • A course with 14 AI-assisted modules that takes you from idea to a production app • Cursor-first workflows that help you build the entire AI application faster and with fewer mistakes Pre-order today and get instant access to: • 15+ AI prompt templates • 40+ Cursor rules • my previous full-stack marketing course (https://www.skool.com/ai-developer-accelerator/classroom). Launch: September 20. Questions? Comment below or DM me.