Hey everyone 👋
I’m currently working with a team that’s developing a new Web3 project around AI + DeFi portfolio management, and we’re now getting into the more serious development stage.
We already have an initial version working, and now we’re discussing how to take it further and make the architecture solid enough for a real production environment.
The main idea is pretty simple:
An AI agent monitors on-chain opportunities, analyzes the user’s portfolio, finds potential optimization opportunities, and then works with predefined user intents to execute strategies on-chain — while keeping the user in control of their assets.
There are a lot of interesting technical questions around this, especially:
How should the AI agent interact with smart contracts?
What should happen off-chain vs. on-chain?
How do we make automated execution safe?
What’s the right architecture for monitoring markets continuously?
How should intents and permissions be designed?
How do we handle failed transactions, slippage, and execution risk?
What needs to change when moving from MVP to production?
So I’d like to hear from people who have actually built things in DeFi, Web3, smart contracts, AI agents, backend infrastructure, or cloud/DevOps.
I’m not really looking for theoretical answers.
I’d rather talk with people who can look at what we’re building and say:
“I’d do this differently here…”
“This could become a problem later…”
“If you want to scale this, I’d build it this way…”
Those kinds of conversations are really valuable for us right now.
If you’re a Web3 developer, smart contract engineer, architect, AI engineer, or just someone with strong experience building DeFi products, feel free to comment or share linkein profile.
Would be great to exchange ideas and hear different perspectives. 🤝
Top comments (0)