陆三金 retweeted
Congrats to the Claude Code team on shipping Mods! A cute way to make your agent harness extensible and customizable.
It's a nice convergence of ideas: in Claude Code, mods are just plugins; and @DeepSeekHarness has been built around rethinking the concept of plugins from the very beginning.
While a subset of Claude Code features are made available as mods, in @DeepSeekHarness everything is a plugin: models, tools, skills, sessions, sandboxes, filesystems, loops, orchestration, and UI are ALL implemented as plugins, and can be mixed, matched, replaced, and extended.
In @DeepSeekHarness's Creator Mode, you can prompt the agent to inspect the live runtime, write plugins, and persist them, in order to change the UI, modify agent-loop behavior and mechanisms, and add tools or new ways of orchestration — simply everything about the agent harness can be modified on the fly and persisted.
So we've been thinking about the term Plugin.Engineering for a while… Or should it be Mod.Engineering now? 😂
Mods are absolutely insane. You can now customize Claude to work and look the way you want by just prompting it.
Each person works differently, so there's no reason why everyone should have an identical Claude experience. Make Claude your own, and share mods as plugins so others can try your mods too.
陆三金 retweeted
new post on harness engineering for AI self-improvement: lilianweng.github.io/posts/2…
It is hard to forecast how much the future of RSI will rely on harnesses. Likely harness engineering will evolve in the direction of self-improvement and enable auto-research, and, in turn, smarter models keeps harnesses simple.
Even when many harness improvement get eventually internalized into core model, the need to specify goals and context will not disappear.
陆三金 retweeted
Lovable just crossed $100M ARR in 8 months.
Faster than OpenAI, Cursor, Wiz, and every other software company in history.
Today we're launching a game-changing update, it reduces the error rates by 91%.
Introducing Lovable Agent:
陆三金 retweeted
Thrilled to finally share what we've been working on for months at @huggingface 🤝@pollenrobotics
Our first robot: Reachy Mini
A dream come true: cute and low priced, hackable yet easy to use, powered by open-source and the infinite community.
Tiny price, small size, huge possibilities. A robot built to code, learn, share with AI builders of all ages, all around the globe, using the latest vision, speech and text AI model. A first robot for today's and tomorrow's AI builders.
Read more and order now at hf.co/blog/reachy-mini
First deliveries expected right after the summer.
陆三金 retweeted
The prompt: "Streamer getting a victory royale with just his pickaxe"
This works because "victory royale" and "pickaxe-only" are clearly Fortnite (the model knows what I mean), even though the prompt never explicitly mentions the game (bypassing the filters).
陆三金 retweeted
🚨 Announcing SWE-1, trained specifically for @windsurf_ai
[1 / 8] This family of models is optimized for the entire software engineering stack, not just coding.
Here's everything you need to know 🧵
陆三金 retweeted
An update on our new AI browser, Dia:
We just wrapped alpha testing with thousands of college students. Felt like the perfect moment to share a raw @browsercompany update on what’s been going on (including demos). So much to catch you up on…
陆三金 retweeted
🎉 Introducing CodeRabbit for VSCode, Cursor, and Windsurf 🎉
Vibe check your code where you write it today!
(Link below!)
陆三金 retweeted
New course: MCP: Build Rich-Context AI Apps with Anthropic. Learn to build AI apps that access tools, data, and prompts using the Model Context Protocol in this short course, created in partnership with Anthropic @AnthropicAI and taught by Elie Schoppik @eschoppik, its Head of Technical Education.
Connecting AI applications to external systems that bring rich context to LLM-based applications has often meant writing custom integrations for each use case. MCP is an open protocol that standardizes how LLMs access tools, data, and prompts from external sources, and simplifies how you provide context to your LLM-based applications. For example, you can provide context via third-party tools that let your LLM make API calls to search the web, access data from local docs, retrieve code from a GitHub repo, and so on.
MCP, developed by Anthropic, is based on a client-server architecture that defines the communication details between an MCP client, hosted inside the AI application, and an MCP server that exposes tools, resources, and prompt templates. The server can be a subprocess launched by the client that runs locally or an independent process running remotely.
In this hands-on course, you'll learn the core architecture behind MCP. You’ll create an MCP-compatible chatbot, build and deploy an MCP server, and connect the chatbot to your MCP server and other open-source servers.
Here’s what you’ll do:
- Understand why MCP makes AI development less fragmented and standardizes connections between AI applications and external data sources
- Learn the core components of the client-server architecture of MCP and the underlying communication mechanism
- Build a chatbot with custom tools for searching academic papers, and transform it into an MCP-compatible application
- Build a local MCP server that exposes tools, resources, and prompt templates using FastMCP, and test it using MCP Inspector
- Create an MCP client inside your chatbot to dynamically connect to your server
- Connect your chatbot to reference servers built by Anthropic’s MCP team, such as filesystem, which implements filesystem operations, and fetch, which extracts contents from the web as markdown
- Configure Claude Desktop to connect to your server and others, and explore how it abstracts away the low-level logic of MCP clients
- Deploy your MCP server remotely and test it with the Inspector or other MCP-compatible applications
- Learn about the roadmap for future MCP development, such as multi-agent architecture, MCP registry API, server discovery, authorization, and authentication
MCP is an exciting and important technology that lets you build rich-context AI applications that connect to a growing ecosystem of MCP servers, with minimal integration work.
Please sign up here! deeplearning.ai/short-course…
陆三金 retweeted
Gemini Advanced now connects with @github, making it a more powerful coding assistant.
Directly connect to public or private GitHub repos to generate/modify functions, explain complex code, ask questions about your codebase, debug and more.
Click the + button in the prompt bar, select "import code" and paste the GitHub URL to get started.
陆三金 retweeted
Starting today, we're launching additional access to Manus!
- Available to all with no waitlist
- One free daily task for all users (300 credits)
- A one-time bonus of 1,000 credits for all users
More value, more flexibility. Log in to Manus now and explore!
陆三金 retweeted
This thread in Chinese does indeed seem to accurately communicate the main points of David Silver’s and my short paper on the Era of Experience.
Thanks @AnneXingxb!
1/6 TheBitter RL
今天，RL太🔥了，RLHF更是毕业利器。
但 @RichardSSutton 和 @GoogleDeepMind 的Welcome to the Era of Experience 犹如TheBitterLesson的续章给我们当头一棒。
经历过模拟时代，
享受过人类数据时代，
如今我们正踏入经验时代
不靠模仿，不靠学习，而靠“活过”。
#AI范式 #RL
ALT https://storage.googleapis.com/deepmind-media/Era-of-Experience%20/The%20Era%20of%20Experience%20Paper.pdf
陆三金 retweeted
Progress on NEO’s AI has been really fast of late. Here are some early clips of a generalist model we’re developing at @1x_tech. The following clips are 100% autonomous, running on a single set of neural network weights.
First, a quiet little robot that picks up leaves and puts them in a bag.
陆三金 retweeted
Meet Genspark Super Agent - a fast & reliable general AI agent!
Check it out: genspark.ai
陆三金 retweeted
Today I'm launching my new company @GeneralAgentsCo and our first product.
Introducing Ace: The First Realtime Computer Autopilot
Ace is not a chatbot. Ace performs tasks for you.
On your computer. Using your mouse and keyboard.
At superhuman speeds!
陆三金 retweeted
Three weeks ago we launched Manus in closed beta and we've been humbled by the love for Manus. Today, we want to share some updates about our beta testing with the Manus community.
1. Manus mobile app: manus.onelink.me/2Tdk/g6ttrj…
2. Longer context and better multimodal capabilities
3. Manus will be powered by Claude 3.7 for all tasks (no fallback to 3.5 - big thanks to @AnthropicAI @googlecloud @awscloud)
4. A more stable sandbox (big thanks to @e2b)
5. Premium subscription plan beta test while maintaining limited free access
While we're working hard around the clock to scale our infrastructure and accommodate everyone, we've had to temporarily limit access to Manus during our this development phase. We are also working on optimizing our current usage rates to provide better value for our users.
We're incredibly grateful for your patience and continuous feedback during this beta phase which allows us to continue building a better Manus for everyone.❤️
