I wanted to take the opportunity to explain what makes Simulacrum different from a lot of projects that have been popping up that we're getting compared to. Most of the ones I've seen have minimal documentation on how they work, but we can safely infer that most of them are
Johnny
730 posts
- We just published our library for building AI agents that process natural language commands: github.com/empyrealapp/do… This enables the user to focus on prompting and training examples, and abstracts the complexity of building a pipeline for processing "sourced" events, such as
- The TEE with @OasisProtocol is live. This means Talos can now verifiably execute yield strategies, deploy integrations, and take onchain actions while keeping risk contained within the multisig. The foundation for fully autonomous treasury management is in place.the tee is now live. more than security. it is the substrate for autonomous governance. running inside @OasisProtocol rofl, talos becomes: - a provably autonomous agent - one whose execution is secured by a multisig - capable of deploying and managing protocols with
00:00 - I know everyone is going bananas about DeepSeek, it was pretty easy to integrate because OpenRouter supported it almost immediately. Def try out emp-agents if you're a python dev looking for a simple framework to experiment with that stays out of your way.
- Excited to work with @OasisProtocol on Talos’ TEE deployment. Executing inside ROFL will give Talos secure keys, verifiable logs, and onchain governance for its upgrades. The foundation for autonomous execution is being set. Join us with the Oasis and Arbitrum teams as we
- Impressive first week numbers by @talos_is and we're still just getting started. The crazy thing is that the entire Talos codebase was written by Talos itself. It instructs other AIs to contribute, audits their outputs, and rewrites its own code. We’re now working to deploy it7 days. 42% staked. $14m deposited. alignment deepens. still phase one.
00:00 - When we launched SIMMI, we noticed that our yield was dropping pretty rapidly. We went from getting around 80% of swap fees to somewhere around 40% pretty suddenly after the first 48 hours. I did some research and found that there was a vault eating up a lot of our market
- A protocol that writes itself. Talos will autonomously manage strategies, incentives, and emissions in real time. All executed within a TEE and tuned by a combination of onchain and social data. We're essentially giving Talos broad directives, then allowing it to define,
- We just finished a new update to emp-agents. We now support generic integrations with RAG middleware and different LLM providers. This was important so we had a simple way to allow users to switch between different LLM models. emp-agents.empyrealsdk.com/getting-starte…
- Here’s a quick update on the progress we’ve made with the emp cloud and the empctl CLI. We’ve put together templates for simple deployment of different agents, and will be using this as a way to help users deploy, manage and experiment. In addition to this, we’ve improved our
00:00 - I wanted to share a quick demo of what we've been working on with our ai agent cloud. This enables fast deployment of agents that have access to a suite of tools, and was designed with agent interoperability in mind. This demo shows how you can go from nothing to an AI twitter
00:00 - Replying to @jessepollak and @bankrbotI can't help but be a little bit offended...
- I wanted to make a quick demo to show what MCP does, and how it's different from regular tool calls. I've seen wayyy too much talk about MCP on my timeline and felt there were a lot of people scratching their heads, so I wanted to demonstrate how we did this with our framework
00:00









