Echo — Data Intelligence Assistant 🚀

Executive Summary

Echo is a state-of-the-art, AI-driven data intelligence platform designed to transform raw tabular datasets into actionable strategic insights. Powered by Google's latest Gemini models (gemini-2.5-flash), Echo provides senior-level data analysis, multi-turn conversational exploration, dynamic Plotly chart generation, automatic Exploratory Data Analysis (EDA), and seamless report exports.

Designed for modern analysts, managers, and decision-makers who demand speed without sacrificing depth, Echo acts as a force multiplier for data-driven teams.

---

🌟 Visual Preview & Key Features

1. Landing & Seamless Data Ingestion

Upload CSV or Excel (.xlsx) files up to standard Streamlit limits with automatic data profiling.

2. Dataset Profiling & Executive Summaries

Instantly inspect row counts, column counts, missing value summaries, numeric column distributions, and detailed previews. Generate structured, C-suite ready executive summaries covering key patterns, risks, opportunities, and strategic recommendations at the click of a button.

3. Automatic Exploratory Data Analysis (EDA)

Echo automatically renders high-level visualizations for uploaded datasets, including numeric distribution histograms, categorical value count bar charts, and interactive Pearson correlation heatmaps.

4. Conversational Intelligence & Dynamic Chart Generation

Ask complex questions about your data with full multi-turn conversation memory. Ask follow-up questions freely (e.g., "expand on that risk", "what about high-income borrowers?"). Request custom charts simply by asking (e.g., "plot a bar chart of loan_amount by homeownership" or "show me a scatter plot"), and Echo writes and executes interactive Plotly visualizations on the fly.

5. Comprehensive Report Export & Easy Copy-Pasting

Export your entire session—including data profiles, the AI executive summary, and complete chat history—as a clean, structured .txt report directly in your browser.

💡 Tip: The exported .txt report is formatted using standard markdown headers, bullet points, and divider sections. You can easily copy and paste its contents directly into Google Docs, Microsoft Word, Notion, or any Markdown (.md) editor without needing any extra formatting!

---

🛠️ Technology Stack

• UI Framework: Streamlit 1.54+
• Data Engine: Pandas & Openpyxl
• Visualizations: Plotly Express & Graph Objects
• AI Core: Google GenAI SDK (google-genai with Gemini models)
• Language: Python 3.12+

---

🚀 Getting Started

Prerequisites

• Python 3.12 or higher.
• A Google Gemini API Key (get one from Google AI Studio).

Installation

1. Clone the repository:

   git clone <repository-url>
   cd echo

2. Set up virtual environment:

   python -m venv venv
   # On Windows:
   venv\Scripts\activate
   # On macOS/Linux:
   source venv/bin/activate

3. Install dependencies:

   pip install -r requirements.txt

4. Configure API Key: Create a .env file in the root directory:

   GEMINI_API_KEY=your_gemini_api_key_here

   Or configure GEMINI_API_KEY inside .streamlit/secrets.toml.

Running the Application

Launch the app with Streamlit:

streamlit run app.py

Open your browser at http://localhost:8501.

---

📖 Usage Guide

1. Upload Dataset: Drag and drop your CSV or Excel file into the file uploader.
2. Review Profile & EDA: Expand Automatic EDA Charts to view data distributions and correlation heatmaps.
3. Generate Summary: Click Generate Executive Summary for an AI analysis of key trends, data quality risks, and opportunities.
4. Chat & Follow-up: Use the Ask Echo chat input to ask questions about your dataset. Ask follow-up questions naturally—Echo retains full conversation context.
5. Request Custom Charts: Ask Echo to plot specific variables (e.g., "show me a distribution of debt_to_income"), and view interactive Plotly charts directly in the chat.
6. Download Report: Click Download Report (.txt) to save your session output, ready to paste into documentation or executive slide decks.

---

Echo — Listen to what your data is telling you. 🔊