Hi everyone,
I’ve spent months building a high-speed intraday trading bot in Python, designed specifically for the Alpaca platform. The project is now roughly 90 percent complete. It’s a serious personal endeavor that has grown to nearly 5,000 lines of code. I’m a Senior Electrical Engineer by profession, so while programming hasn’t been the central focus of my career, I do have significant technical experience. The bot is designed to be strategy-agnostic, meaning it can support virtually any intraday trading strategy with minimal changes to the core system.
One of the main advantages of the program is its flexibility. It allows for easy testing or trading of a wide variety of strategies. I’ve built several proof-of-concept modules that demonstrate specific features, and most of those features are now fully integrated. What remains is refining the final integration, fixing bugs, and verifying that all features work as intended when combined. The bot is built for paper trading on Alpaca, but once fully functional, transitioning to a live account should be very simple.
The program uses the SIP WebSocket connection that provides tick data with nanosecond-precision timestamps. The bot is therefore designed to process data and execute trades at subsecond speeds, with a target of under 300 milliseconds for most client-side executions. It’s built to handle data and execution faster than any human day-trader could react.
The tech stack includes a broad range of libraries: asyncio, aiohttp, websockets, requests, pandas, numpy, pickle, dateutil, pytz, socket, logging, threading, cachetools, and more. Development started in Visual Studio Code and has since moved to Cursor, but I’m open to using whichever IDE works best for collaboration.
Although the program is currently not modularized in the strictest sense, many of the proof-of-concept modules were successfully integrated into a single, cohesive system. I’m open to re-modularizing the structure if it would simplify things and help with debugging or future development.
The remaining ten percent of work is mostly debugging and aligning all the logic to ensure every part behaves according to its intended function. I’m also open to new ideas or enhancements if my collaborator has suggestions. Right now, the focus is not profitability but simply getting the bot to function reliably with all its base features.
I’m looking for a collaborator with a formal education in computer science, software engineering, or a similar field—or someone with several years of professional programming experience, particularly with Python. If you have experience in real-time systems, algorithmic trading, or event-driven architecture, that’s a plus. In return for your help, you’ll have full access to the source code for your own personal use, under the condition that it not be shared, sold, or distributed. We can sign a simple contract to protect both our interests. Once the system is functioning properly, we can talk about opportunities to sell a standalone version if there’s interest, but that’s not the focus right now.
I have also prepared a high-level scope of work describing the design and functionality of the system, along with a flowchart that illustrates how the entire program is structured. I’d be happy to share those documents with anyone seriously interested in partnering.
If this sounds like something you’d like to be part of, feel free to reply here or send me a direct message. I look forward to hearing from you.