Strategy Copilot

What the company appears to be: Strategy Copilot is a self-reported tool that allows users to describe trading strategies in natural language (e.g., "10/50 day MA crossover on AAPL since 2022") and receive automated backtesting results, including equity curves and performance metrics. It was built by one developer, Mike Zhang, using Codex (a generative AI tool) as a coding partner.

What changed: The project is described as a personal development effort for an intern preparing for quant trading roles, with the goal of building a practical backtesting tool while testing AI-assisted development workflows. No evidence of prior version or evolution beyond this single submission.

Single most important open question: Is there any evidence that Strategy Copilot has been used by actual users or deployed in production? The description lacks any mention of customers, revenue, or adoption — only a developer's account of building it.

The description states that Strategy Copilot is a tool that allows users to describe trading strategies in plain English and returns full backtests including equity curves, total return, max drawdown, and buy-and-hold comparisons. It supports two strategy types: moving average crossover and RSI threshold. It also includes side-by-side comparison mode, parameter sweep functionality, and typo-tolerant natural language parsing.

The author reports that Codex was used to write core logic (signal logic, parser, strategy comparison functions, unit tests), while the developer drove architecture and stress-testing.

Evidence: The project description.

Confidence: Low — this is self-reported, unverified functionality.

The product is positioned as a tool that turns natural language trading instructions into automated backtesting. It claims to support international tickers, be typo-tolerant, and order-independent in parsing.

There is no indication of prior versions or evolution beyond this single submission. The author’s own write-up frames it as a learning exercise for quant trading internships and a test of AI-assisted development.

Evidence: The project description.

Confidence: Low — the claim is self-reported, with no external validation or prior product history.

The description does not state who the target customer is. It only mentions that the author is preparing for quant trading internships and wanted to build a useful tool. No evidence of actual users, customer segments, or personas is provided.

Evidence: The project description.

Confidence: Not evidenced — no mention of target customers or ICP.

There is no evidence in the description of any business model or pricing structure. The author describes building a tool for personal and internship use, with no indication of monetization or commercial intent.

Evidence: The project description.

Confidence: Not evidenced — no mention of pricing, revenue, or business model.

The product is built using Codex (a generative AI tool), Python libraries like pandas, matplotlib, yfinance, and streamlit. It includes a strategy-agnostic backtest engine, supports two strategy types, and has a parameter sweep feature.

The author reports that the natural language parser was stress-tested with real-world phrasing and edge cases, including ticker truncation bugs and Yahoo Finance rate-limiting issues.

Evidence: The project description.

Confidence: Low — this is self-reported technical implementation, not verified in production.

There is no evidence of traction or adoption. The author states that the tool was built for personal development and internship prep. No users, customers, or usage data are mentioned.

The project is described as a single submission to a hackathon, with no indication of prior versions or commercial deployment.

Evidence: The project description.

Confidence: Not evidenced — no signs of traction or maturity.

No competitive analysis or market context is provided in the description. The author does not mention competitors or similar tools in the quant trading or backtesting space.

Evidence: The project description.

Confidence: Not evidenced — no mention of competitive landscape.

• Unverified claims: All functionality and features are self-reported, unverified.
• No traction or adoption: No evidence of users, customers, or real-world usage.
• Single-person development: The project was built by one person, with no indication of team or support structure.
• Hackathon submission: The tool is described as a hackathon entry, suggesting it may not be production-ready or commercially viable.
• No monetization or business model: No evidence of how the product would generate revenue.

Evidence: The project description.

Confidence: Low — these are inferences based on absence of evidence.

1. What is the actual use case for this tool? Is it intended for personal use, or is there a plan to commercialize it?
2. Has anyone else used Strategy Copilot beyond yourself? If so, what feedback have you received?
3. Are you planning to monetize this product? If yes, how?
4. What are the technical limitations of the current version that would prevent it from being production-ready?
5. How do you plan to scale beyond a single developer?

Evidence: The project description.

Confidence: Low — these are questions for clarification, not based on evidence.