Data Interpreter: An LLM Agent For Data Science

- achieves top performance in ml and reasoning, analyzing stocks /replicating websites. it autonomously debugs code and solves issues using notebooks and browsers.
- os

https://github.com/geekan/MetaGPT/tree/main/examples/di

GM, time for DeepMind to drop another revolutionary breakthough ♥️
https://deepmind.google/discover/blog/sima-generalist-ai-agent-for-3d-virtual-environments/

The missing pieces of AI agents: vast goals, state buildup, high skill in risky tasks, missing features, autonomous goals, and hidden operation.

https://www.lesswrong.com/posts/6BB9yqXKYjwnajYpM/does-increasing-the-power-of-a-multimodal-llm-get-you-an

Scaling Protocol Intelligence with Web3 Agents…

today by @namesty Polywrap at #AgentsUnleashed

Can I bill warps multiple times in the same frame interaction?

One of our biggest predictions I’d that in the next 5-10 years, everyday people will get access to a trained AI agent that will help them navigate their health

Ie. It does the same or better job as status quo (an overworked doc coming in and talking to you for 10-15 minutes)

"We show that LLM agents can autonomously hack websites, performing tasks as complex as blind database schema extraction and SQL injections without human feedback. Importantly, the agent does not need to know the vulnerability beforehand." 😅 https://t.co/Bkc2gupDWL

Dad, musician, Tech geek, musician, business manager, and all round curious dude.

I’ll start: AI Agent crypto enabling systems. Why? AI agents are the perfect use case for crypto given the digital nature of these agents, and if you consider the immense value they may create for humanity in the future they will need rails to transfer value easily, there will be no higher value use case in crypto.

Lots of interesting early experiments of AI agents in crypto including:
- NPC consumer app agents (frenrug)
- Smart wallets with AI agents (Dawn)
- Gaming agents like Parallel's Colony game
- AI agent marketplaces 

What other types of AI agents are people excited about? 
https://nystrom.substack.com/p/crypto-ai-agents

Coded an agent that scans my YouTube subscriptions for new videos, thanks them by relatedness to the product I am building right now, and suggests ideas from those videos

Is anyone working on large action models (LAMs) to actually introduce AI and the “autonomous” part into DAOs? Would love to talk to anyone building action or decision models.

https://x.com/yoheinakajima/status/1756046363140506030?s=46

More Agents Is All You Need
☺︎ 
⇩
finds the performance of llms improves w/ more agents using a simple sampling-and-voting technique. this approach, effective across different tasks, adds to existing methods for enhancing llms. (note: its effectiveness varies w/task difficulty)
https://arxiv.org/abs/2402.05120

Looking into providing L2 liquidity for CODE to make it easier to join D_D. Any L2 you’d prefer to see? Why?

“We introduce SELF-DISCOVER, a general framework for LLMs to self-discover the task-intrinsic reasoning structures to tackle complex reasoning problems that are challenging for typical prompting methods.”

https://arxiv.org/abs/2402.03620

AnyTool: Self-Reflective, Hierarchical Agents for Large-Scale API Calls

⇩

https://arxiv.org/abs/2402.04253

Large Language Model based Multi-Agents: A Survey of Progress and Challenges

↓
https://arxiv.org/abs/2402.01680

1. It would be crazy if Vision Pro collected public data to improve its system? Like street view for example could benefit hugely from this and so would apple maps. But that's just the tip of the iceberg.

Over long term, Apple would have information about how humans interact with surroundings. Contd in comments

Founding AI Engineer at AutoGPT | Former Steward @developerdao

Nick Tindle

@ntindle