Sebastian Lehrer pfp

Sebastian Lehrer

@sebastianlehrer

147 Following
348 Followers
Sebastian Lehrer pfp
Sebastian Lehrer
@sebastianlehrer
Tokenized collectibles is one of the most interesting use-cases for NFTs I've come across with.
0 reply
0 recast
0 reaction
Sebastian Lehrer pfp
Sebastian Lehrer
@sebastianlehrer
Your onchain reputation will become much more important a few years from now.
0 reply
0 recast
1 reaction
Sebastian Lehrer pfp
Sebastian Lehrer
@sebastianlehrer
Hey Tim - I've sent you a DC.
0 reply
0 recast
1 reaction
Sebastian Lehrer pfp
Sebastian Lehrer
@sebastianlehrer
My takeaways from an interview to @tldr about /bracket and consumer apps: The approach you take when building a consumer app is not the same if you're building for crypto-native people vs non-crypto-native people. These apps should enable new behaviors—regardless of who the end user is. Building a consumer-grade app for non-native users involves tradeoffs (for example, not supporting WalletConnect and forcing the user to create a wallet from scratch). Every product team should decide when and if their users should realize they are actually using a crypto app. Adding social capabilities to an app makes it much more fun, but the social part doesn’t have to be the core of the app. A balance needs to be achieved. And a great point from @dawufi: Crypto onboarding should be much better. If you wanted to show a friend how cool the internet is, you’d show them Wikipedia, Google, or Facebook. In crypto, the first thing you show them is how to set up a wallet.
2 replies
1 recast
11 reactions
Sebastian Lehrer pfp
Sebastian Lehrer
@sebastianlehrer
Hey man - would love to interview on thepodcaster.xyz and have you explain what Pixie is about. Are you interested in that? :)
0 reply
0 recast
0 reaction
Sebastian Lehrer pfp
Sebastian Lehrer
@sebastianlehrer
Is Openrank labelling accounts as spam based on some sort of algorithm? Or is it going to be more like a community thing in which user A can label user B as a spam?
0 reply
0 recast
0 reaction
Sebastian Lehrer pfp
Sebastian Lehrer
@sebastianlehrer
Testing from the app? 👀
1 reply
0 recast
1 reaction
Sebastian Lehrer pfp
Sebastian Lehrer
@sebastianlehrer
Takeaways from an interview that @@benfielding from @gensyn recently gave. Listen to the full conversation here https://www.youtube.com/watch?v=g4jPsXMs7Gw
0 reply
0 recast
0 reaction
Sebastian Lehrer pfp
Sebastian Lehrer
@sebastianlehrer
This means that a potentially dangerous model could be blocked at the application level (rather than much earlier, before the model is even created, as current regulations attempt to do).
1 reply
0 recast
0 reaction
Sebastian Lehrer pfp
Sebastian Lehrer
@sebastianlehrer
(iii) Auditability – Open sourcing ML models makes it much easier to understand why someone is building something and what they are trying to achieve with a specific model.
1 reply
0 recast
0 reaction
Sebastian Lehrer pfp
Sebastian Lehrer
@sebastianlehrer
That radically changes the sort of ML models we can create. Your compute power is no longer limited to Google data centers but potentially includes every phone and compute-having device.
1 reply
0 recast
0 reaction
Sebastian Lehrer pfp
Sebastian Lehrer
@sebastianlehrer
(ii) Increased scale – You essentially remove the scaling limit. If we can only run training/inference operations on data centers, we face huge constraints in terms of the compute power available.
1 reply
0 recast
0 reaction
Sebastian Lehrer pfp
Sebastian Lehrer
@sebastianlehrer
(i) Lower prices – If we create an open market for ML compute (i.e., enable anyone to provide CPU/GPUs for the training/inference of these models), the increase in supply will lower costs and solve the huge logistics burden we currently face.
1 reply
0 recast
0 reaction
Sebastian Lehrer pfp
Sebastian Lehrer
@sebastianlehrer
In the future, apps and websites will replace imperative code execution with probabilistic ML models, and we will get to a point where we will have intelligent agents acting on our behalf. This will obviously significantly increase the demand for compute power.
1 reply
0 recast
0 reaction
Sebastian Lehrer pfp
Sebastian Lehrer
@sebastianlehrer
So, why do we want to decentralize this space? What do we get out of this? 3 things - (i) lower prices, (ii) increased scale, (iii) auditability.
1 reply
0 recast
0 reaction
Sebastian Lehrer pfp
Sebastian Lehrer
@sebastianlehrer
Machine Learning is an inevitable tech, essentially because of how effective it is in multiple areas. The internet as we know it today will broadly be replaced by machine learning models.
1 reply
0 recast
0 reaction
Sebastian Lehrer pfp
Sebastian Lehrer
@sebastianlehrer
Why the world needs a Machine Learning Compute Protocol (i.e., @gensyn) 🧵
1 reply
0 recast
0 reaction
The PodCaster 🎤 pfp
The PodCaster 🎤
@thepodcaster
#8: Meet Wildcard Tune in to our eighth episode, where we interviewed @defikaran.eth from /wildcardclub. We talked about how Wildcard was inspired by Friend Tech and Degen, their plans to onboard Crypto Twitter, his thoughts on tipping as a meta, the Wildcard mobile app, and much more https://www.youtube.com/watch?v=06rlLnJE_8U
2 replies
2 recasts
8 reactions
Sebastian Lehrer pfp
Sebastian Lehrer
@sebastianlehrer
Tipping is a game-changer feature for social apps. Imagine if creators would get a cent each time they get a like on Twitter or an upvote on Reddit. No creator is making money out of their social pressence, all their revenue comes from ads/selling merch. Tipping changes that.
0 reply
0 recast
0 reaction
Sebastian Lehrer pfp
Sebastian Lehrer
@sebastianlehrer
Huge
0 reply
0 recast
3 reactions