Inspiration
Every trader has faced moments of regret---whether it was selling too early, holding too long, or taking unnecessary risks. I wanted to create a tool that helps traders learn from their past mistakes, analyze their decisions, and improve their future trades with AI-powered insights.
What it does
TradeRewind is an AI-powered trading assistant that:
- Analyzes past trades to detect mistakes, such as early exits or high-risk decisions.
- Forecasts stock trends using historical data and machine learning models.
- Simulates alternative scenarios to show how different decisions could have impacted profits.
- Clusters trading behavior to classify users into different trader types and provide tailored advice.
- Provides AI coaching with personalized tips to refine trading strategies.
How I built it
- Frontend: Streamlit for a simple, interactive UI.
- Data Processing: Pandas and Scikit-Learn for trade analysis.
- Stock Data: Yahoo Finance API for real-time and historical stock data.
- AI & Forecasting: Prophet for time-series forecasting and OpenAI's API for trading insights.
- Hosting: Streamlit Cloud with a custom domain (TradeRewind.tech).
Challenges I ran into
- Ensuring accurate trade mistake detection based on diverse trading behaviors.
- Implementing real-time financial data integration without hitting API rate limits.
- Choosing and implementing a machine learning model for clustering and forecasting
Accomplishments that I'm proud of
- Successfully built a trading mistake analyzer and stock price forecasting in a limited timeframe.
- Integrated AI-driven financial coaching to help traders improve decision-making.
- Created an interactive data visualization dashboard to analyze trade performance.
What I learned
- The importance of behavioral analysis in trading, beyond just numbers.
- How small decision changes can significantly impact financial outcomes.
- Best practices in deploying AI-powered financial applications securely.
What's next for TradeRewind
I plan to expand TradeRewind with more advanced features to make it a comprehensive trading assistant:
- Social hub for day traders : A community feature where traders can share insights, discuss strategies, and learn from each other's mistakes.
- Learn about different strategies: An educational section to explore various trading techniques, risk management strategies, and real-world case studies.
- Geopolitical & Sentiment Analysis: AI-powered insights that analyze global events, news sentiment, and market trends to predict trading risks.
- Risk Assessment Dashboard: A more detailed risk analysis tool to assess trade volatility, diversification levels, and overall portfolio health.
- Enhanced AI Trading Coach:More personalized insights, integrating real-time data and predictive modeling for better decision-making.
- Mobile-friendly version: A lightweight mobile app or responsive UI for easier access on the go.
Built With
- machine-learning
- openai
- prophet
- python
- sklearn
- streamlit
- yfinance


Log in or sign up for Devpost to join the conversation.