Slowly but surely github.com/ai16z/eliza/pu…

Here is a diagram of the system I've been working on this week. At first, it will be used to allow agents to experience emotions similarly to how we do. Later, this can be expanded to allow agents to feel based on inputs from sources beyond messages as well. I know a lot of the

Been afk for a bit. Had holidays, intel processor issue, then FYE busyness at work. Excited to dive back into projects next month and catch up with the changes.

I got bored and released a concept album. If you like glitchy, fast-paced, weeb music, have at it. open.spotify.com/album/5JNwD3A0… on.soundcloud.com/sbtD1Gkmpj0RIa…

There's something interesting happening in crypto again. Lots of aspiring streamers entering the space with high work ethic but low crypto experience. Launching coins with little to moderate success. Not sure what to make of this.

Last year I earned some certs for work, published an iOS app, and started a business. This year, I took a vacation for the first time in my life. I still live in a freezing hellscape where no one shares any of the interests I hold. I live two lives. In one, I surgically

I actually love watching the trenches without buying anything.

I made this plugin for ElizaOS. Although consistent behavior can be ideal for agents, I was hoping to have an agent that acts based on an updating emotional state rather than a static persona. Emotional state can be altered by external factors (think weather, harsh words from

Gave Clawdbot access to my portfolio. "Trade this to $1M. Don't make mistakes" 25 strategies. 3,000+ reports. 12 new algos. It scanned every X post. Charted every technical. Traded 24/7. It lost everything. But boy was it beautiful.

Ah. So building a trading agent is expensive.

claude code is fucking insane i know literally NOTHING about coding. ZERO. and i just built a fully functioning web app in minutes. http://localhost:3000/ check it out

Seems like people want their agents to just be models.

I realize now that I’m not a coder. I’ll be having not me write lovecraftian horror novels going forward. github.com/Junsgh/digital…

Random thought: Why do agent frameworks need plugins if they already have knowledge of their environment? They can just run programs pretty well now right?

My agent wrote a book. Not technically impressive, but cool.

I think elizaOS was so close to OpenClaw🦞 but differences I noticed as a dev/user: 1. openclaw bootstrapped more fun things for "normies". Eliza could do the things too, but took effort from the dev to find/create the plugin and troubleshoot if unstable. 2. openclaw

Models are the tech. Agents are the tools. It's the same loop that drove Moore's Law. We built tools to make smaller chips, then used those chips to make better tools. AI is entering that same recursive cycle. Every agent action is a potential training example. If your agent is

This app runs a youtube channel by itself. It doesn't use AI generated video. It selects footage based on relevance to the script and edits for proper timing. The web ui is chopped but it's not my priority. Autonomous too. github.com/boyaloxer/scri…