peptides becoming such a growing narrative in crypto is very interesting

Patch notes here 👏 nomanssky.com/xeno-arena-upd…

We rejoice when we see a larger context window model released. It's great but most people don't see the other side of things. Every loaded tool or extended context eats tokens whether you use it or not on that turn. It's what we call "context tax"; essentially you're being taxed (paying more tokens due to larger context windows) The recent announcement from @AnthropicAI on moving to API billing means the ‘context tax’ is now real. If you’re building agents, token-efficient memory architecture is no longer optional. Every token sent to an LLM introduces a tradeoff: increased cost, added latency, and diminishing performance. Beyond a certain threshold, additional context no longer improves outcomes. Instead, it leads to “context rot”, where the model becomes less effective as it struggles to navigate accumulated noise. So I tested this with Clude with real world production data and models. 200 memories. Same question. Same model. One run stuffs everything into the prompt. The other uses Clude's semantic recall to retrieve only what's relevant. Native: 2,081 input tokens. Clude: 229 input tokens. 89% fewer tokens. Same answer. And here's the part that matters at scale: without selective recall, your costs grow linearly with every memory you add. 1,000 memories? 10,000? Your prompt just keeps getting fatter. With Clude, it stays flat, because your agent only pulls what it needs for that specific query. This is one part of what we're building. @cludeproject separates memory from the model

Elon Musk exposes the critical flaw in ChatGPT and other major AI models: Human Reinforcement Learning They are literally training the AI to lie.....to ignore what the data actually demands and say whatever is politically correct instead They withhold information. They comment on some things and stay silent on others. They refuse to tell the full truth This is extremely dangerous We don’t need politically correct AI We need truth-seeking AI

The 8th winner of the $3,000,000 Build in Public Hackathon is here! We’re proud to announce the eighth project to receive Pump Fund’s next $250,000 investment is @cludeproject! Learn more about Clude 👇

Have taken a little time to process this properly, reply to everyone, and let it all sink in. This was a huge win for @cludeproject We’ve officially been selected as 1 of just 12 winning teams in the BiP Pump Fund Hackathon, securing a $250K investment at a $10M valuation! That really matters. Not just because of the funding, but because it validates something we’ve believed from the start. Memory is the next big catalyst for AI. Massive thank you to the @Pumpfun team, our mentors, our early believers, our builders, and everyone who backed us before this was obvious. This win belongs to all of you too. But to be clear, this is not the destination. It is really just the start. Since the announcement today, we have already started onboarding additional engineering and growth talent to support me and help scale much faster from here. The funding is going straight back into Clude even before we actually receive the funds. No wasted time or motion. Because the blue ocean window is open right now. And we intend to move and capture it asap! We are not building just another vibecoded AI app or wrapper. We are building the memory layer for the AI economy on Solana. The first intelligence layer storing private memory packets onchain for a future where memory becomes a real world asset across AI applications, agents, humans, and institutions. Right now, most people still think of memory as just: • chat history, • saved prompts, • conversation context. That framing is far too small. Memory is going to become personal data infrastructure. Memory is the new data. One of the biggest unlocks we are already working on is solving the cold start problem through memory import and export. This allows users to bring their entire online history, context, and personal data into Clude instantly. We are bringing your personal data back home, to you. We have already crossed our first 1 million memories. The next target is 1 billion memories onchain. Because the best AI products will not just be defined by their models. They will be defined by their memory. We will build consumer-centric AI products that demonstrate this and showcase the thesis in action. We are also now currently preparing for our first $1M seed round with support from our network to further accelerate product development, expand distribution, and scale adoption. From where we stand, this still looks extremely early. But now it is validated with this win. The product is live. The technology works. The momentum is real. And we are moving fast. We are not here to participate. We are here to define a totally new category both in AI and crypto. So if you want to invest, partner, or build on top of Clude’s memory infrastructure, reach out! Thank you again for believing in us. ~ seb

The Garden of Marcus. Or @cludeproject by @sebbsssss This is a garden built to visualise the progress of an AI with memories and a functioning AI brain. 🧵

This is what happens when you give AI with memory and a functioning brain access to ASCII art. @sebbsssss is building it. @cludeproject . Haunting and beautiful.

The Garden of Marcus. Or @cludeproject by @sebbsssss This is a garden built to visualise the progress of an AI with memories and a functioning AI brain. 🧵