Update README and FAQ: Refine descriptions and add usage instructions for the chatbot

HUGGINGFACE_SOLUTIONS.md

@@ -1,61 +0,0 @@
-# Hugging Face Spaces File System Solutions
-
-## The Problem
-Hugging Face Spaces have a **read-only file system**, meaning your app cannot write to CSV files or any files on disk. This is expected behavior for security and resource management.
-
-## Current Quick Fix (Implemented)
-✅ **Session State Storage**: The app now stores user details and unknown questions in Streamlit's session state:
-- Visitors can still leave their contact details
-- You can see collected leads in the sidebar during their session
-- Data persists for the duration of their browser session
-
-## Permanent Solutions for Production
-
-### 1. **Database Integration** (Recommended)
-Connect to an external database to store contact details permanently:
-
-**Options:**
-- **Supabase** (PostgreSQL, free tier available)
-- **MongoDB Atlas** (NoSQL, free tier available)  
-- **Google Sheets API** (Simple, visual)
-- **Airtable API** (User-friendly interface)
-
-### 2. **Email Notifications**
-Instead of storing data, send immediate email notifications:
-```python
-# When user provides contact info, send email via:
-# - SendGrid API
-# - Mailgun API  
-# - SMTP (Gmail, etc.)
-```
-
-### 3. **Webhook Integration** 
-Send data to external services via webhooks:
-```python
-# Send to:
-# - Zapier webhook
-# - Make.com (Integromat) 
-# - Custom API endpoint
-```
-
-### 4. **Google Sheets Integration**
-Simple solution using Google Sheets as database:
-```python
-import gspread
-# Write directly to Google Sheets
-# Easy to view and manage leads
-```
-
-## Implementation Priority
-
-1. **Immediate** ✅: Session state storage (already done)
-2. **Short-term**: Email notifications for new leads
-3. **Long-term**: Full database integration
-
-## Next Steps
-
-1. Test the current Hugging Face app - errors should be gone
-2. Choose a permanent storage solution
-3. Set up email notifications for immediate lead capture
-
-The current fix ensures your chatbot works without errors while you decide on the best permanent solution.

README.md

@@ -11,40 +11,29 @@ tags:
 - openai
 - data-solutions
 pinned: false
-short_description: AI Assistant v1.0.3 - Blue Bean Data services and solutions
+short_description: AI Assistant - Blue Bean Data services and solutions
 ---
 
-# Blue Bean Data AI Assistant ☕
+# Blue Bean Data AI Assistant
 
-An intelligent chatbot powered by OpenAI that helps visitors learn about Blue Bean Data's services and capabilities.
+>A minimal AI-powered chatbot for Blue Bean Data, built with Streamlit and OpenAI.
 
-## Features
+## What it does
 
-- 🤖 **AI-Powered Conversations**: Uses OpenAI GPT-4o-mini for natural language understanding
-- 📚 **Knowledge Base**: Includes comprehensive FAQ and team information
-- 📊 **Contact Recording**: Automatically records interested visitors' details
-- 🔍 **Unknown Question Tracking**: Logs questions that need human follow-up
-- 💬 **Natural Interface**: Clean, professional chat interface
+- Answers questions about Blue Bean Data's services using a built-in FAQ and OpenAI GPT-4o-mini.
+- Records visitor contact details for follow-up.
+- Tracks unknown questions for later review.
 
-## Services We Cover
+## How to use
 
-- Data Strategy & Consulting
-- BI & Dashboards
-- Data Pipeline Development
-- AI & Automation
-- Predictive Modeling & Analytics
-- Database Design & Optimization
+1. Start the app (Streamlit, Docker, or Hugging Face Space).
+2. Ask questions about Blue Bean Data's offerings or expertise.
+3. The chatbot responds instantly and can collect your contact info if you wish.
 
-## Usage
+## Requirements
 
-Simply start a conversation by asking about Blue Bean Data's services, expertise, or how we can help with your data challenges!
-
-## Setup
-
-This application requires an OpenAI API key. Set it as a secret named `OPENAI_API_KEY` in your Hugging Face Space settings.
-
-## Version History
-- v1.0.3 (2025-01-20): Blue theme, session-based storage for HF compatibility
+- Python 3.10+
+- OpenAI API key (set as `OPENAI_API_KEY` environment variable)
 
 ---
-Built by Blue Bean Data - Transforming data into business value.
+Made by Blue Bean Data.

src/chatbot_data/faq_blue_bean_data.md

@@ -54,6 +54,19 @@ A: This service is for companies struggling with slow application performance, d
 Q: What technologies do you work with?
 A: We use a modern and robust tech stack, including Python, SQL, AWS, Google Cloud, Azure, Databricks, Snowflake, Power BI, Tableau, Streamlit, Scikit-learn, TensorFlow, OpenAI, and more.
 
+Using the Chatbot
+Q: How do I use the Blue Bean Data chatbot?
+A: Simply type your question or message into the chat interface. The chatbot will respond instantly with helpful information based on our FAQ, knowledge base, and AI-powered understanding. You can ask about our services, team, process, technologies, or anything related to Blue Bean Data.
+
+Q: What kind of questions can the chatbot answer?
+A: The chatbot is designed to answer a wide range of questions, including:
+- Details about Blue Bean Data's services and expertise
+- Information about our team and company background
+- Our approach, process, and technologies used
+- How to get in touch or start a project
+- General data, analytics, and AI topics
+If your question is outside the chatbot's knowledge, it will log it for human follow-up, so you always get the help you need.
+
 Our Approach & Process
 Q: What is your process for working with new clients?
 A: We follow a transparent, five-step process:
@@ -88,4 +101,19 @@ Q: How soon can you start a new project?
 A: Our contact form includes an option to indicate your timeline, whether it's "Immediately," "Within a month," "Within 3 months," or "Just exploring options." We will discuss a specific start date during our initial conversations.
 
 Q: How does your pricing work?
-A: We work on a project basis. After our free hands-on discovery session, we provide a detailed project proposal that outlines the scope, deliverables, and associated costs, so you have a clear understanding of the investment before we begin.
+A: We work on a project basis. After our free hands-on discovery session, we provide a detailed project proposal that outlines the scope, deliverables, and associated costs, so you have a clear understanding of the investment before we begin.
+
+Q: Do you offer freelance or contract-based services?
+A: Yes, we also provide freelance and contract-based data services for clients who need flexible, short-term, or specialized support. Whether you need extra hands for a specific project, expert advice, or ongoing part-time help, we can tailor our engagement to fit your needs. Contact us to discuss your requirements and how we can help as freelance data professionals.
+
+Q: Can you build a custom chatbot like this for my business?
+A: Absolutely! The chatbot you are interacting with was built using modern technologies like Streamlit and OpenAI's GPT models. If you are interested in a similar AI-powered chatbot for your own website or business, we can design and implement a tailored solution to fit your needs, including integration with your data and branding. Contact us to discuss your requirements and get a proposal.
+
+Chatbot Technical Implementation
+Q: How was this chatbot built?
+A: This chatbot was developed using a combination of modern AI and web technologies:
+- **OpenAI GPT models** for natural language understanding and response generation
+- **Streamlit** for the interactive web interface
+- **Langchain** and other GenAI frameworks for advanced prompt engineering, retrieval, and workflow orchestration
+- Deployed on **Hugging Face Spaces** for easy public access
+We can adapt the technical stack to your needs, including integration with your own data sources, APIs, or cloud infrastructure.