Liquid Arena

InspirationInspiration

Crypto trading today is fragmented. Traders constantly switch between multiple tools—price trackers, news feeds, charting platforms, and exchanges—just to answer simple questions like what to buy, whether now is a good entry point, and how to execute a trade quickly. We were inspired to simplify this experience by building a conversational AI trading copilot that brings discovery, analysis, and execution into a single interface.

What We Built We built Liquid Arena, an AI-powered trading assistant that allows users to interact with the market through natural language and voice. Users can ask questions about crypto opportunities, check prices, get AI-generated market analysis, set alerts, and execute trades directly from the chat interface. The system aggregates market data, technical signals, and sentiment insights to generate actionable recommendations and streamline the trading workflow.

How We Built It The project uses a Next.js frontend for the chat interface and dashboard, paired with a FastAPI backend that handles intent parsing, market data retrieval, and trade logic. We integrated speech recognition and text-to-speech so users can speak to the agent and hear responses. For the MVP we used in-memory storage for strategies and trades, with the architecture designed to support a persistent database such as Supabase. Market data and analysis modules power the AI’s recommendations and alerts.

What We Learned We learned how powerful conversational interfaces can be when combined with structured financial data. Building the system forced us to think carefully about translating natural language into concrete actions like querying markets, setting conditions, and placing orders. We also gained experience designing modular backend services that allow AI reasoning, trading logic, and UI interactions to work together smoothly.

Challenges We Faced The biggest challenge was integrating multiple components—AI intent parsing, real-time market data, chat interactions, and voice capabilities—within a short hackathon timeframe. Coordinating frontend and backend development in parallel required clear API contracts and rapid iteration. Another challenge was balancing automation with safety, ensuring that trade actions require explicit confirmation and that the AI’s recommendations remain transparent to the user.

Overall, this project showed us how AI can transform complex financial workflows into simple conversations, making trading faster, more intuitive, and more accessible.

What it does

How we built it

Challenges we ran into

Accomplishments that we're proud of

What we learned

What's next for Liquid Arena

Built With

• claudecode
• liquid
• react