DEV Community

Sachin Kumar Das
Sachin Kumar Das

Posted on

TradeMemory

Project Documentation: TradeMemory

Exploring Memory-Augmented AI for Trading Journaling

Tech Stack: MERN + Groq (Qwen-3) + Hindsight Cloud Vector SDK


Overview

While working on AI systems and full-stack development, we explored how Large Language Models (LLMs) can be combined with vector databases to create applications with persistent context.

TradeMemory is a learning project focused on building an intelligent trading journal that stores not only trade details but also the reasoning and emotions behind trading decisions.

The goal was to experiment with how AI memory can improve the way users review past experiences.


The Problem

Many retail traders maintain their trading records using spreadsheets or basic notes.

While these methods store information, they often lack:

1. Easy Retrieval

A normal spreadsheet cannot understand questions like:

"What was my mindset when I made my previous loss on Zomato?"

2. Consistent Logging

Manual journaling often becomes difficult to maintain because entering multiple fields after every trade creates friction.

3. Learning From History

Stored data usually remains static and does not provide context from previous decisions.

TradeMemory explores a different approach by allowing users to log trades naturally through conversation and storing those interactions as searchable memories.


System Architecture

The application follows a simple flow:

Recall → Process → Store

1. Semantic Recall (Hindsight SDK)

Before generating a response, the system searches previous stored memories using vector similarity.

Relevant past experiences are retrieved and provided as context.

2. Context Processing (Groq / Qwen)

The retrieved information is passed to the LLM so the response can consider previous interactions.

3. Memory Storage

After the interaction, important trade-related information is identified, tagged with metadata, and stored for future retrieval.


AI Processing Approach

The project uses Qwen-3 through the Groq API.

A custom agent layer manages:

  • Conversation history
  • Prompt structure
  • Response formatting
  • Memory extraction

Regex-based filtering is used to keep responses clean and focused for the user.


Features Implemented

1. Natural Language Trade Logging

Instead of filling multiple fields manually, users can write:

"Booked 20k profit on Tata Motors, felt disciplined today."

The system identifies the intent and stores it as a structured trade memory.


2. Trading Statistics Dashboard

The dashboard displays:

  • Total trades
  • Wins
  • Losses
  • Win rate

The statistics are generated from stored memory metadata.


3. Trader Reflection Model

The system can analyze previous trade memories and generate a summary of observed patterns.

Example:

"User performs better with large-cap stocks and struggles during volatile setups."


Security Implementation

API keys are handled only on the backend.

The frontend communicates through a Node.js middleware layer, preventing direct exposure of sensitive credentials.


Technical Details

Frontend

Built using React.js.

The interface uses:

  • Dark terminal-style UI
  • Monospace typography
  • Quick action prompts

React hooks are used to refresh statistics after new memories are stored.


Backend

Built using:

  • Node.js
  • Express.js

The backend manages:

  • LLM communication
  • Memory retrieval
  • Data processing

The official Hindsight SDK is used for vector operations.


Vector Database Concept

Instead of storing only text, each trade memory is converted into an embedding.

This allows similarity-based searching.

For example:

Searching "Reliance trade" can also retrieve memories containing:

  • RIL
  • Blue chip investment
  • Previous similar setups

because they have similar meaning.


Challenges During Development

Metadata Handling

Initially, dashboard values were not updating correctly.

This was improved by adding structured metadata tags during memory creation.

Response Speed

Using Groq inference improved response time significantly compared to traditional LLM API calls.

AI Boundaries

Prompt constraints were added to keep the system focused on journaling and analysis rather than providing financial advice.


Future Improvements

Possible extensions:

Automated Trade Import

Connecting with broker APIs to import trade history.

Voice Logging

Allowing users to record trade thoughts through voice.

Market-Specific Memory

Separate memory contexts for:

  • NSE
  • US Markets
  • Crypto

Conclusion

TradeMemory was built as an experiment to understand how persistent AI memory can be used in real applications.

The project combines:

  • Full-stack development
  • LLM integration
  • Vector databases
  • Prompt engineering

to explore a more contextual way of interacting with AI systems.

Top comments (0)