A collection of AI/ML projects exploring generative models, NLP, and database interactions. Includes Streamlit apps for real-world use cases.
A Streamlit application that allows users to ask questions about SQL databases using natural language. Leverages Groq LLM for query generation and LangChain for database interaction.
- Real-time SQL query generation
- Schema-aware responses
- Error handling for database interactions
- Context-aware conversations
- Python | Streamlit | LangChain | Groq LLM | MySQL
git clone https://github.com/vivekpathania/ai-experiments
cd sqlchatbotVirtual Environment:
pip install virtualenv
python3 -m venv env
source env/bin/activateInstall dependencies:
pip install -r requirements.txtCreate .env file: env
DB_URI=mysql+pymysql://username:password@localhost/dbname
GROQ_API_KEY=your_groq_api_key_hereRun the app:
streamlit run sqlchatbot.pyAutomatically evaluate candidate fit for job roles using Groq LLM and Streamlit. Generates match scores, personalized improvement suggestions, and ATS-compliant CV rewrites.
- Score-Based Recommendations:
- Score ≥8: CV is aligned; no changes needed.
- 5 ≤ Score <8: Receive improvement suggestions and CV rewrite.
- Score <5: Redirected to refine or find a better-fitting role.
- Real-Time Analysis: Instant feedback on CV-JD alignment.
- ATS-Compliant CV Rewrite: Structured for applicant tracking systems.
- Loading Spinners: Visual feedback during LLM processing.
- Dynamic UI: Tailored experiences based on evaluation scores.
- Python | Streamlit | LangChain | Groq LLM | PyMuPDF
git clone https://github.com/vivekpathania/ai-experiments
cd hrappVirtual Environment:
pip install virtualenv
python3 -m venv env
source env/bin/activateInstall dependencies:
pip install -r requirements.txtRun the app:
streamlit run app.py- Upload Files: Drag-and-drop PDFs for the job description and CV.
- Select Mode: Choose between Hiring or Candidate Mode.
- Evaluate Fit: Click "Evaluate Fit" to generate a report.
- Improve CV: Review suggestions and generate an updated CV (Score ≥5).
This project is licensed under the Apache License.
A conversational travel planning application that generates customized itineraries using AI. Supports both business and leisure trips with tailored recommendations.
- Conversational Interface: Natural language interaction to gather trip details
- Real-Time Data Integration: Uses Tavily/SerpApi to fetch up-to-date travel information
- Customized Itineraries: Generates Markdown-formatted itineraries with flight options, hotel recommendations, and daily schedules
- Multi-Tool Support: Integrates Groq LLM, Tavily, and SerpApi for comprehensive data processing
- Visual Formatting: Uses emojis, tables, and bullet points for clear presentation
- Python | Streamlit | Agno | Groq LLM | Tavily/SerpApi
git clone https://github.com/vivekpathania/ai-experiments
cd travel-agentVirtual Environment:
pip install virtualenv
python3 -m venv env
source env/bin/activateInstall dependencies:
pip install -r requirements.txtRun the app:
streamlit run app.py- Enter API Keys: Provide Groq and Tavily/SerpApi keys in the sidebar
- Describe Your Trip: Use natural language to specify trip details (e.g., "Family holiday from London to Paris for 4 days")
- Generate Itinerary: The AI will create a detailed itinerary with options for flights, hotels, and activities
Business trip from New York to Tokyo from 2024-05-10 to 2024-05-15 for 1 traveler. Budget: $3000. Needs conference venues and after-work dining options.# **Tokyo Business Trip Itinerary**
## Trip Summary
- **Type:** Business
- **Dates:** May 10 - May 15, 2024
- **Travelers:** 1
- **Budget:** $3,000
## Flight Options
| Airline | Price (Adult) | Details |
|------------------|---------------|--------------------------|
| JAL | $1,200 | Non-stop, 8 AM departure |
| ANA | $1,150 | 1 layover in Seoul |
## Accommodation
| Hotel | Price/Night | Amenities |
|----------------------|-------------|--------------------------|
| Park Hyatt Tokyo | $350 | Business center, WiFi |
| Mandarin Oriental | $290 | Fitness center, breakfast|
## Daily Schedule
- **May 10**: Business meeting at Conference Center X (10 AM)
- **May 11**: Networking dinner at Robata-tei (7 PM)
## Important Tips
- **Transportation**: Use the Tokyo Metro for easy access to business districts.
- **Culture**: Familiarize yourself with Japanese business etiquette.This project is licensed under the MIT License.