X-Alpha
Find the Trends, Be Early and Get Profit
Created on 15th May 2025
•
X-Alpha
Find the Trends, Be Early and Get Profit
The problem X-Alpha solves
Retail crypto traders are overwhelmed by noise and lack actionable intelligence. While platforms like Twitter, Telegram, and on-chain explorers are rich in alpha, most users don’t have the tools to filter signal from noise, identify opportunities early, or take action instantly. Existing solutions either focus on price charts or general analytics—none combine real-time social intelligence with onchain execution in one seamless flow.
X-Alpha.ai bridges the gap between insight and action.
We’ve integrated Coinbase’s AgentKit to allow users to perform onchain actions (like buy/sell or approve) directly from within our platform. But we go a step further—we combine it with X-Alpha’s proprietary social analytics engine to surface the why behind token movements.
Our AI Assistant now does 3 key things:
Surfaces early signals using Twitter and Telegram data: trending tokens, KOL mentions, follower spikes, and narrative heatmaps.
Guides user decisions through conversational insights, helping them understand social momentum and risk signals before acting.
Enables instant execution via Coinbase’s AgentKit—no need to switch tools or miss timing.
This unlocks a first-of-its-kind trading flow:
Discover → Validate → Execute — all within one unified interface.
By combining context, clarity, and execution, we solve the core UX pain of crypto trading today.
Challenges we ran into
There are several challenges/obstacles that we ran into:
-
Integrating Coinbase AgentKit with our existing backend
Coinbase’s AgentKit is powerful but designed with a specific structure in mind. Our backend (Python/Flask) had to be extended and adapted to handle wallet delegation, action simulation, and secure execution without compromising our existing stack.
We overcame this by modularizing the integration and extending AgentKit's action provider with several other tools (including the Xalpha APIs/MPCs) along with Langgraph integration for graph based workflow. -
Mapping social insights to onchain actions
X-Alpha’s intelligence layer deals primarily with social signals (mentions, trends, influencer activity), while AgentKit expects clear onchain intents (buy, sell, swap).
Bridging this gap was non-trivial—we had to create an AI-driven decision layer that translates social alpha into actionable trading suggestions/actions.
- Real-time data noise filtering
With thousands of social posts every hour, surfacing the right tokens at the right time—without false positives (least amount of false positives)— is still a major challenge.
✅ We built layered filters combining smart follower tracking, narrative clustering, and mention velocity thresholds to ensure our alerts were both early and reliable.
We are also still employing AI and LLMs to verify authenticity of authors/tweets/content on Twitter and on-chain data to provide even more accurate insights.
Tracks Applied (5)
AI
Stablecoins
Consumer
Showcase
DeFi
Technologies used
Cheer Project
Cheering for a project means supporting a project you like with as little as 0.0025 ETH. Right now, you can Cheer using ETH on Arbitrum, Optimism and Base.