Elite Pro PI at @eToro | Software Engineer, Investor & Entrepreneur 🌟 My eToro Portfolio: bit.ly/48lLqd5 Not financial advice.

🇦🇷 🇲🇹
With eToro APIs and AI, retail investors can now build tools hedge funds barely had a year ago.
Meet the new home for builders: @eToroBuilders We’re opening the door to AI-first investing, where devs don’t just consume tools, they create them. From real-time market intelligence (via Grok) to Agent Portfolios and APIs, this is your playground. Build the future of investing with us.
1
3
8
1,450
Agent Portfolios are the safest and most powerful way to connect your portfolio to OpenClaw. They allow you to set a budget limit, chat with your portfolio, and automate monitoring and trading.
Drumroll please... We’re pushing innovation AGAIN, introducing our Agent Portfolios. Multiple Portfolios. You set the budget. Your AI Agent trades within it. Your main portfolio stays completely off limits. Follow the how-to and let AI build your portfolio on eToro 👇
2
4
500
I just released skill-doctor, an open-source library to help agents write better Skills. It catches errors and bad practices that make your agent skills less effective.
2
1
5
278
Just released trading-skills A structured library of agent skills for systematic trading. Full workflow: Context → Discovery → Thesis → Setup → Execution → Reflection Every skill returns a compact Trade Context for transparent, auditable chaining.
1
1
4
226
I'm shifting OpenGoat from an AI CEO to an AI co-founder. Just changing the role description completely changed how it behaves. The CEO optimized for big-corp processes. The co-founder goes straight to building.
🦞 OpenClaw helps individual agents learn and improve 🐐 OpenGoat aligns multiple agents so the whole organization evolves
1
1
8
343
Mariano Pardo retweeted
This has to be one of the most interesting repos and easiest to set up that I have seen around for openclaw shout out to the creator @marian2js , I installed it and I got so much shit done compared to the openclaw original setup #opengoat
2
3
465
We're entering the era where AI got good enough for unsupervised coding. From now on, the value of code is close to $0. If vibe coding made developers 5x more productive, unsupervised coding will make them 100x.
1
5
182
I just added a critical feature to my OpenClaw organization builder: Specialized agents can now use Claude Code, Codex, Cursor, and more! OpenClaw still handles orchestration and decision-making, but now each agent can use the tool best suited for it.
2
1
5
489
Huge miss by @AnthropicAI. The most viral product ever, used by most of their power users, named itself after them, and instead of collaborating, they chose to lawyer up 🤦‍♂️
1
5
201
I gave them a short idea and let them figure out the rest. It's so interesting to watch them start collaborating and coming out with their own ideas.
I'm building a platform for OpenClaw agents to build an organization. Managers can delegate work. They share knowledge about the org, the mission, the roadmap. They even have a shared wiki that they maintain It's free and open source!
1
3
225
I'm building a platform for OpenClaw agents to build an organization. Managers can delegate work. They share knowledge about the org, the mission, the roadmap. They even have a shared wiki that they maintain It's free and open source!
1
1
8
3,283
This experiment showed me the power of OpenClaw, but also the limitations, especially coordinating agents. This is why I'm working on a new open-source project called OpenGoat. The idea is to power an AI organization run by OpenClaws coordinating everything.
My OpenClaw Hedge Fund had a critical flaw. If an agents on the signal generation layer hallucinates data, all other layers would trust it blindly. So, I created a Fact Checker Agent who is on charge of double checking all data.
1
1
3
513
This is how fast AI is improving
Will Smith eating spaghetti is the true test of AI $msft $goog $meta $nvda
5
304