DP
DHRUV PATEL
AI & DATA SCIENCE · GEC RAJKOT
STATUS
● ONLINE
DATE
VERSION
v 2.6
SYSTEM BOOT PROGRESS0%
INITIALIZING SYSTEMS...
PORTFOLIO v2.6
Back to Projects
Artificial Intelligence

Nexa AI — Smart Chatbot

A customizable AI chatbot built with Streamlit, LangChain, and DeepSeek/Groq. Features login/signup, saved chat history, Shayari/Joke/Quote modes, Lottie animations, and Render deployment.

Interactive Showcase Gallery (3 Slides)

Nexa AI — Smart Chatbot - Intelligent Agent
Slide 1 of 3

Intelligent Agent

Project Specifications

CategoryArtificial Intelligence
Statuscompleted
Author / ArchitectDhruv Patel

Ready to explore this project in action?

Check out the live interactive deployment or inspect the full source code on GitHub.

Technologies Used

Python 3.9+StreamlitLangChainDeepSeekGroq APILottie FilesdotenvRender

About The Project

Nexa AI is a next-generation conversational AI assistant built with Python, Streamlit, and LangChain. It provides a fun, practical, and culturally flavored chatbot experience with support for both DeepSeek and Groq LLM APIs.

Beyond basic chat, Nexa has built-in custom response modes: Gujarati Jokes, Shayari (Urdu poetry), Chanakya Quotes, and Motivational Quotes — switchable from the sidebar. The app includes a full user login/signup system (auth.py), saved and downloadable chat sessions, and Lottie animations for an engaging welcome screen.

The modular architecture separates concerns cleanly: bot.py (core chat logic), auth.py (session management), sidebar.py (navigation), custom_responses.py (special modes). API keys are managed via .env with dotenv. The app is deployed on Render (not just localhost) with a live production URL.

This project demonstrates real LangChain orchestration (not just direct API calls), user state management in Streamlit (which requires creative session handling), and clean Python software architecture.

Key Features & Architecture Highlights

LangChain orchestration with DeepSeek R1 and Groq API support

User login/signup system with session state management in Streamlit

Save, download, and delete chat sessions from the sidebar

Custom response modes: Gujarati Jokes, Shayari, Chanakya Quotes, Motivational Quotes

Lottie animations for welcome and onboarding screens

Modular Python architecture: bot, auth, sidebar, and custom responses separated

Secure API key management via .env with dotenv

Deployed on Render with a live production URL

Interested in building something extraordinary?

Whether you have an ambitious AI project, full-stack application, or collaboration in mind, I'd love to connect.