Colin Stephenson@colin_stephenson139
1
Events attended
1
Submissions made
Guyana
1 year of experience
About me
I’m a Full Stack Developer and passionate about building practical technology that solves real-world problems. I hold a B.S. in Computer Science from the University of the People and am currently pursuing an M.S. in Cybersecurity at NYU Tandon School of Engineering. My experience spans full-stack web development, AI-powered applications, automation, and cybersecurity. I enjoy turning ideas into working products and exploring how AI can improve decision-making, workflows, and everyday life. I have worked on projects including Aethyron, an experimental autonomous coding agent; Alphora, an AI-focused hackathon project; MatchMind; and Aetheris. I’m particularly interested in emerging technologies, intelligent systems, and the intersection of software development and cybersecurity. My goal is to keep building innovative solutions that have practical impact while continuously expanding my technical skills.
🤓 Latest Submissions
Alphora
Alphora AI Trading Agent is an MCP-powered financial technology platform that enables AI assistants to interact with a simulated trading environment through structured, machine-readable tools. The core of the project is a Rust-based Model Context Protocol server that exposes Alpaca's Trading and Market Data APIs to AI agents. It provides 23 tools covering the complete paper-trading lifecycle, including account information, portfolio history, positions, order management, assets, stock and ETF market data, cryptocurrency data, options chains, and watchlists. The system is designed to bridge the gap between conversational AI and financial APIs. Instead of an AI assistant merely providing textual market analysis, an MCP-compatible agent can retrieve current market information, inspect portfolios, manage watchlists, and submit or cancel simulated orders through explicit tools. Safety is a central design consideration. The default configuration targets Alpaca's paper-trading environment, allowing autonomous-agent workflows to be developed and tested using simulated capital rather than real funds. The project also includes a React-based web interface providing a visual trading and portfolio experience, with components for account information, positions, order history, order entry, portfolio and price charts, watchlists, and options-chain exploration. The backend is implemented in Rust using the official MCP Rust SDK, asynchronous Tokio runtime, Reqwest for API communication, Serde/Schemars for structured data and schema generation, and Axum/Tower HTTP infrastructure. The frontend is built with React and Vite. Alphora demonstrates how Model Context Protocol can be used to transform an AI assistant from a conversational interface into a controlled, tool-using agent capable of interacting with real-world financial APIs in a sandboxed environment.
4 Sep 2026
