عمر.eth

@omarperacha.eth

56 Following

53 Followers

عمر.eth pfp

@omarperacha.eth

Definitely not in Dubai anymore ⛈️

0 reply

0 recast

1 reaction

عمر.eth pfp

@omarperacha.eth

"Mathematicians have long studied how shapes can fit together to cover surfaces without gaps. However, their typical approach – using shapes with sharp corners and flat faces – is rarely seen in the natural world. Instead, living organisms use a dazzling array of patterns to form and grow, for instance in muscle tissues. Most strikingly these patterns are characterised by shapes with curved edges, non-flat faces, and few, if any, sharp corners. Up to now, how nature achieves geometrical complexity using these ’soft shapes’ has eluded mathematical explanation. The answer... is a new class of mathematical shapes called soft cells. These shapes have a minimal number of sharp corners, and cover space without gaps." https://www.ox.ac.uk/news/2024-09-12-mathematicians-discover-new-universal-class-shapes-explain-complex-biological-forms#:~:text=A%20team%20of%20mathematicians%20from,sea%20shells%20to%20muscle%20cells.

0 reply

0 recast

2 reactions

عمر.eth pfp

@omarperacha.eth

Biggest blockchain conference in the world but I still have to look after a badge AND wristband for a week 😂 “Don't lose your badge or wristband. Treat these with the same level of care as you would a passport.” 🤦🏽‍♂️ Proof of registration in a mobile wallet wen?

0 reply

0 recast

0 reaction

عمر.eth pfp

@omarperacha.eth

Can you give a bit more background on what you did? Learning new crypto business models is a good kind of learning 😎

0 reply

0 recast

0 reaction

عمر.eth pfp

@omarperacha.eth

Overall, it’s just a really well-written paper. Easy to read and very open about shortcomings. These are discussed in insightful detail. Well worth a read if feeling inspired, otherwise there is a blog post version along with fully open-sourced code too 👌🏽 https://sakana.ai/ai-scientist/

0 reply

0 recast

0 reaction

عمر.eth pfp

@omarperacha.eth

The AI Scientist works by providing a set of specified topics, and providing some starter code for each. For example, you might give “language modelling” as a general research topic, and provide Karpathy’s NanoGPT as starter code. It then uses an existing LLM to generate research ideas on the topic, pick one it deems novel, and edits the starter code to implement its idea and run experiments. Afterwards, the same existing LLM writes up the results in a paper using LaTeX and Python for plotting charts. If you wanted to provide a custom topic, you’d need to provide that starter template code too. Naturally, it really only lends itself to software-based research atm.

1 reply

0 recast

0 reaction

عمر.eth pfp

@omarperacha.eth

As background, I spent six years working in generative AI before becoming disillusioned with the resource requirements for the next generation of breakthroughs. I moved to crypto in part because I thought it’s still possible for a small team to do really meaningful work here. But research into agent-based AI application could be a really fruitful new frontier for science. I didn’t really appreciate this until reading this paper

1 reply

0 recast

0 reaction

عمر.eth pfp

@omarperacha.eth

Rather than train a new massive model, the team creates an “agent” that uses and instructs existing LLMs and coding tools to do the work. Think of it as a command line AI “app”, in the same way that Perplexity is a web app which uses existing models like Claude and GPT4 as a core part of the workflow

1 reply

0 recast

0 reaction

عمر.eth pfp

@omarperacha.eth

Just got round to reading “The AI Scientist” paper by Sakana AI, along with teams in Oxford and Canada What I found most exciting about this work is that, unlike most recent AI breakthroughs in the current “scaling war” era, this one really could come from a team with modest resources. https://arxiv.org/abs/2408.06292

1 reply

0 recast

1 reaction

Raluca pfp

@nosoulforsale.eth

Paired.World takes SocialFi to the next level by integrating real-world engagement. Connect, earn, and grow both online and offline. Say hello to the future of social networks!

0 reply

1 recast

2 reactions

عمر.eth pfp

@omarperacha.eth

https://onchain100-frame.vercel.app/api last week I introduced hybrid-chain NFTs as an effective way to decentralise dynamic NFTs on Ethereum, using /icp today I'm open-sourcing IC canister code to get started with your own hybrid-chain NFTs! https://github.com/omarperacha/eth-metadata-starter Example on @base:

0 reply

0 recast

1 reaction

عمر.eth pfp

@omarperacha.eth

I'll open-source a starter project for an ICP canister that read metadata from EVM, so devs can easily explore making their own hybrid NFTs In the meantime, try minting your own from the frame and choosing your favourite colour and icon from the tech projects that made this possible!

0 reply

0 recast

0 reaction

عمر.eth pfp

@omarperacha.eth

Some considerations: · Canisters on IC spend Cycles to make HTTP requests. Each call to this canister's getMetadata endpoint costs about 1.5m Cycles, or roughly $0.0002 · IC canister are upgradable - to ensure decentralised metadata, the canister developer should remove the controller so its code can't be edited

1 reply

0 recast

0 reaction

عمر.eth pfp

@omarperacha.eth

The result - dynamic NFTs whose traits are still searchable on all your favourite marketplaces and by rarity indexers, even after they change. We had this challenge with our dynamic SOUL tokens at PairedWorld, and hybrid chain NFTs seem like a good solution just remember to refresh metadata on OpenSea after a change!

1 reply

0 recast

0 reaction

عمر.eth pfp

@omarperacha.eth

When the canister receives a request from the contract via the tokenURI method, it reads the on-chain trait values directly from the actual contract and generates a string formatted as a JSON conforming to the Metadata standard and returns it as the HTTP response

1 reply

0 recast

0 reaction

عمر.eth pfp

@omarperacha.eth

ICP canisters can receive and make HTTP requests. The premise of this hybrid-chain dynamic NFT is pretty simple The NFT is from a standard ERC-720 contract on @base, but the baseURI is set to the url of an ICP canister

1 reply

0 recast

0 reaction

عمر.eth pfp

@omarperacha.eth

Dynamic NFTs have a dilemma: ・Either you centralise metadata to make traits easy to change ・Or you store traits in the contract, but lose indexing by not returning them from the tokenURI Until now, thanks to /icp Free customisable (!) mint on @base to explore for yourself https://onchain100-frame.vercel.app/api

1 reply

1 recast

3 reactions

عمر.eth pfp

@omarperacha.eth

Real. World. Interactions. 👏

0 reply

0 recast

1 reaction

عمر.eth pfp

@omarperacha.eth

PSA 🚨 if you’ve been building an iOS dApp with self custody, and banging your head against the wall because your users can’t even connect their wallet without the flow breaking: Zerion have just fixed the deep linking issue on affecting ALL users of ALL wallet apps on iOS 17+ Thanks @evg and team ✌🏽

0 reply

0 recast

0 reaction

عمر.eth pfp

@omarperacha.eth

What’s the easiest way to do the following on ICP: Given a canister written in rust, call a method declared in its candid API via HTTP, passing arguments as query params? Do you need a frontend canister or can you directly call this rust “backend” canister?

0 reply

0 recast

0 reaction