Stop Using Claude Chat. Do This Instead.
The video explains the different components of Claude AI and how they relate to each other, arguing that users should avoid Claude Chat and instead use Claude Code over the simpler Cowork version. The speaker recommends using the desktop version and working with local folders to build persistent AI teams.
Summary
The speaker addresses confusion among their coaching members about Claude's different versions and components. They start by explaining the browser version, which includes basic chat functionality and projects for providing static context, but argue this approach is too limiting because conversations start from scratch each time. The speaker advocates for building an AI team with persistent memory instead. They then discuss Claude Code in the browser, noting that non-technical users often avoid it due to intimidating terminology, and explain that the browser version of Code is particularly confusing with unclear repository selection options. The main recommendation is to use the desktop version of Claude, which provides three tabs: chat, code, and cowork. Claude Code allows users to select local folders instead of repositories, making it more intuitive. The speaker explains that Cowork is essentially a simplified version of Claude Code designed to attract broader audiences - it provides visual task lists and file displays that make Claude's work more transparent, but it's less stable and crashes frequently. Claude Code offers the same functionality with more reliability and features. They also mention a terminal version that provides raw input/output with maximum capabilities. Additional components discussed include a Chrome extension that works with websites but lacks persistent context, and a mobile version that only includes chat and Claude Code (no Cowork). The speaker emphasizes that Claude Code's ability to work with local folders enables remote control through mobile devices, allowing users to continue work from anywhere. They briefly mention dispatch functionality and Excel/PowerPoint integrations as additional features.
About this episode
Claude has 7 different versions and most people use the wrong one. Here's the only overview you need. 🚀✨ Build your Future-Proof AI Productivity System TODAY: https://myicor.com In this video, I break down every component of Claude so you actually understand what to use, what to skip, and how they all connect. No features deep dive, just a clear, holistic overview in under 10 minutes. If you've been confused by Claude Chat, Claude Code, Cowork, the Chrome extension, the mobile app, and everything in between, this is the video that clears it all up. Timestamps: 0:00 Why Claude is confusing 0:25 Claude in the browser (Chat) 1:06 Why I don't recommend Chat 1:14 Building an AI team with persistent memory 1:35 Claude Code in the browser (confusing) 2:23 Download the Desktop App 2:38 Claude Code on Desktop (local folders) 3:09 Plan mode and persistent memory 3:28 Claude Cowork explained 3:51 Why Code is better than Cowork 4:44 The takeaway so far 5:03 Claude Code in the Terminal (raw version) 5:30 Chat vs. Code: the simple breakdown 5:53 Chrome Extension 6:23 Using Chrome Extension through Code/Cowork 6:47 Claude Mobile App 7:06 Mobile as remote control for Code 7:27 Why Claude Code wins 7:47 Excel and PowerPoint integrations #Claude #ClaudeCode #ClaudeAI #ClaudeCowork #Anthropic #AIProductivity #AITools #ProductivitySystem
Key Insights
- The speaker claims that most of their inner circle members find Claude confusing during coaching sessions, which motivated the creation of this explanatory video
- The speaker argues that Cowork is essentially Claude Code with a simplified interface designed to attract broader audiences, but Claude Code is more reliable and stable while Cowork crashes frequently
- The speaker explains that the Chrome extension becomes more powerful when used through Claude Code because it can access full context from local folders rather than starting conversations from scratch
Topics
Transcript
[0:00] If there's one thing that we learn from our inner circle members when we have the coaching sessions is that Claude is confusing and this is why I'm making this video. So you will understand in less than 10 minutes, What the different components of Claude actually are. I'm not diving into Claude features. that I did already. In other videos. In this video, you will just get a holistic understanding Of, the different versions of Claude that you can use. Let's dive right into this. Let's start with the browser version. That's maybe the version that you start with usually, and in the [0:32] browser version, this is what you see. You can create a new Chat and…
Full transcript available for MurmurCast members
Sign Up to AccessMore from ICOR with Tom | AI Productivity
Claude sucks. And here is why.
This video explores why Claude and other AI models produce inconsistent results, examining how model selection (Haiku vs Sonnet vs Opus vs Claude 3.5 Fable), effort levels, prompt clarity, and system context dramatically affect output quality. The speaker demonstrates these differences through a Venn diagram creation task and advocates for using organized folder systems with documented SOPs and code-based solutions to ensure consistent AI performance.
Claude can now prompt its own SESSIONS. I built the boss.
Claude now has the ability to interact with multiple sessions, allowing a streamlined orchestration of tasks through an agent called Larry. This integration can simplify task management and enhance productivity by enabling a single main session to oversee various sub-sessions or tasks.
I switched off Claude's memory. My folder does it better.
Tom explains why he disabled Claude's native memory feature and instead built a local folder-based system (myPKM) to maintain persistent context with AI agents. He shares an audit revealing systemic issues like contradictory documentation, forgotten rules, and parallel session conflicts, then outlines how to systematically fix these problems using the ICOR productivity methodology.
I do not download invoices anymore. My folder and Claude do it.
The speaker demonstrates how he uses AI (Claude) integrated with his local folder system to automate invoice collection for his accountant, downloading 36 invoices in 5 minutes with minimal manual intervention. He explains his AI team methodology, work streams, and how persistent local knowledge bases enable AI to handle recurring tasks independently without constant re-explanation.
Claude finished the work. Then I lost it in the chat.
A creator demonstrates a comprehensive local-folder-based system for managing AI-generated deliverables asynchronously, featuring an interactive browser interface that tracks decisions, changes, and conversations without relying on external databases or web applications. The system enables collaborative editing, visual annotations, and automatic context preservation across multiple parallel AI projects.