Back
Year
2026
Tech Stack
Python, Streamlit, pandas, NumPy, SciPy, Plotly, LangChain, Groq LLM
Description
A Streamlit chatbot for conversational analysis of CSV and Excel
files. A LangChain and Groq pipeline translates plain-English questions into
Pandas, SciPy, and Plotly operations, enabling data profiling, missing-value
detection, outlier detection, correlation analysis, and visualisation without
requiring users to write code.
Why This Project Matters
The project is designed to reduce the manual exploratory data
analysis workflow for non-technical users working with tabular data. It connects
natural-language interaction with actual statistical and data-analysis operations
rather than limiting the chatbot to descriptive responses.
Technical Highlights
- Built a LangChain + Groq pipeline that maps natural-language questions to Pandas, SciPy, and Plotly operations
- Implemented automated correlation analysis and outlier detection through plain-English queries
- Added automatic data profiling on file upload without requiring manual configuration
- Built a unified ingestion workflow supporting both CSV and Excel files
Key Features
- CSV and Excel file upload
- Automatic data profiling
- Missing-value and outlier detection
- Correlation analysis
- Natural-language querying with visualisation generation
My Role
- 🧩 Worked as the sole developer across the full application
- 🧠 Built the LangChain + Groq pipeline for mapping natural language to Pandas and SciPy operations
- 📊 Implemented automatic data profiling, missing-value analysis, and outlier detection
- 📈 Built the Plotly visualisation generation layer
- 🗃️ Built the CSV/Excel ingestion and validation workflow and designed the Streamlit chat interface end to end