avatar
CryptoNerdCn 🦇🔊
12,323 Twitter followers
Follow
Delegate of @ensdomains | 中文&日本語&English | @WasmCairo | Into @starknet by @starknetastrocn
Posts
avatar
CryptoNerdCn 🦇🔊
Thread
Google's Gemini has fallen far behind GPT in both search and de-illusion capabilities, even lagging behind manual Google searches. I truly didn't expect Google's own AI to stumble in search. The cause: I asked Gemini if they had an obsidian image recognition plugin. My requirement was simple: to convert images in my notes into text using a right-click menu or command. Results: 1. Gemini provided a completely unrelated plugin and persisted with its de-illusion answer even after I followed up with multiple questions, attributing the problem to messy project documentation and release history 😂 (I was on the Pro plan, and the conversation wasn't in fast mode). 2. I asked ChatGPT with the exact same prompts, and it worked perfectly on the first try. All four of the suggested plugins were usable, each with a link for verification on the ChatGPT page. 3. Finally, I conducted an experiment, manually searching "obsidian image recognition plugin" on Google, and immediately found a usable plugin. I can only say that I'm glad my Gemini Pro plan uses the company's plugin. I would never pay for this myself.
GPT
3.54%
avatar
CryptoNerdCn 🦇🔊
In other words (or, practically speaking): Is your main job a marketing account (like those so-called AI gurus on WeChat who just copy and paste English information)? If not, first, block all those marketing accounts and only look at information from official product sources. Then, consider what aspects of your work and life you can use AI to handle. After trying one or two things, congratulations, you're already enjoying the AI era.
马东锡 NLP
@dongxi_nlp
人的注意力是有限的。 能用好claude code，codex 以及一款国产AI产品，再有点精力读读论文其实就已经到极限了。 不要过度浪费精力在乱七八槽的东西上，除非你，“想，并且，能” ，靠这个东西赚钱。 但大多数人不能，且会被剥夺注意力。 x.com/dotey/status/2…
avatar
CryptoNerdCn 🦇🔊
02-06
Thread
A quick summary of Claude Code's Agent teams: A lead session can initiate multiple independent teammate sessions for parallel collaboration. They are self-organized through a shared task list and messaging system, making them more suitable for discussing, exploring, and collaborating on complex tasks. Key features include: - Difference from subagents: Subagents involve a lead agent assigning tasks, with subagents only reporting results; agent teams allow multiple teammates to communicate directly, share task lists, and claim tasks themselves, making them more suitable for complex tasks requiring discussion and collaboration. - Activation method: Disabled by default; requires setting the environment variable `CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS=1` (configurable in `settings.json`). - How to start: Use natural language to have Claude create a team and describe your desired team structure (roles/number of members/division of labor). After creation, a shared task list will be generated. The lead is responsible for aggregation and completion, and you can interact directly with each teammate. - Presentation Modes: Two UI modes - In-process: All team members run within the same terminal, switching between concurrent messaging via Shift+Up/Down. - Split panes: Each team member has their own pane (requires tmux or iTerm2 + it2 CLI/Python API), default is auto. - Specify team members and models: Claude can decide the number of teammates, or explicitly require (e.g., "4 teammates refactor in parallel") and specify the model used by each team member. - Plan approval: Team members can be required to submit a plan in read-only mode before the lead approves it and begins coding; approval criteria can be provided in the prompt (e.g., tests must be included). - Delegate mode: Allows the lead to only orchestrate without writing code directly (Shift+Tab toggles), suitable for scenarios where you want to turn the lead into a pure orchestration. - Task Assignment and Preemption: The shared task list supports pending/in-progress/completed tasks and dependencies; leads can assign tasks, and teammates can also claim them themselves; file locks prevent multiple people from competing for the same task. - Shutdown and Cleanup: Leads can request a teammate to shut down a task; upon completion, the lead can perform a cleanup (checking for active teammates to avoid resource residue/inconsistency). - Local Storage and Architecture: Team configurations and task lists are stored locally (e.g., `~/.claude/teams/{team}/config.json`, `~/.claude/tasks/{team}/`), with a mailbox for message delivery and idle notifications. - Token Cost: Teams significantly increase token consumption (each teammate's independent context), suitable for research/review/new features; daily small tasks are more cost-effective per session. - Usage suggestions and troubleshooting: Emphasizing "providing sufficient context, appropriate task splitting, avoiding file conflicts, and continuous monitoring and guidance"; listing common problems (teammates not appearing, too many permission prompts, lead ending work too early, orphaned tmux, etc.). twitter.com/cryptonerdcn/statu...
loading indicator
Loading..