Building Data Apps with Streamlit

Turn Python Scripts into Interactive AI-Powered Apps with Streamlit

Key Features
● Build complete Streamlit apps from data exploration to deployment
● Integrate ML models and AI chatbots into interactive Python apps
● Master caching, state management, cloud deployment, and multipage architecture

Book Description
Streamlit has transformed how developers present data science and machine-learning work by making it effortless to turn Python scripts into interactive web applications. Building Data Apps with Streamlit provides a complete, hands-on roadmap to creating professional, production-ready apps using Streamlit’s fast, intuitive, and Pythonic framework.

You begin with Streamlit’s architecture, layout system, and component ecosystem, learning how to build clean, scalable apps with widgets, callbacks, caching, and session state. The book then guides you through handling secrets, managing configurations, working with APIs and databases, and building multipage workflows that feel polished and responsive.

By the end, you will build a full Streamlit solution that analyzes datasets, trains machine-learning models, and powers an AI chatbot using Google Gemini. With dedicated chapters on testing, optimization, and cloud deployment, this book equips you with the confidence and skills to create, iterate, and share high-quality Streamlit applications that bring your data and ideas to life.

What you will learn
● Build interactive data apps using Streamlit’s core components
● Manage session state, caching, themes, and configurations effectively
● Connect apps to APIs, databases, and cloud services
● Integrate machine-learning workflows into Streamlit interfaces
● Create and deploy an AI chatbot using Google Gemini
● Test, deploy, and maintain Streamlit apps on the cloud

1. Introduction to Streamlit
2. Setting Up the Development Environment
3. Creating and Deploying Your First Streamlit App
4. Exploring Streamlit’s Flow and Architecture
5. Persisting Data and State Across App Reruns
6. Exploring Streamlit’s Page Elements
7. Widget Keys and Callbacks
8. Introduction to Streamlit Caching and Connections
9. Managing Secrets in Streamlit
10. Advanced App Management Concepts
11. App Configuration Options
12. Building Multipage Streamlit Apps
13. Testing Streamlit Apps
14. Building a Data Analysis Streamlit App
15. Building a Machine Learning Streamlit App
16. Building a Chatbot on Streamlit
Index

Siddhant Sadangiis a Developer Experience Engineer at neptune.ai, focusing on Python, MLOps, and Streamlit. A recognized Streamlit Creator and Community Moderator, he builds open-source tools and guides developers in turning data science projects into interactive web apps.

About the Technical Reviewer
Siavash Yasini is an ex-physicist turned full-stack data scientist and machine learning practitioner. As a technical lead, he works at the intersection of data science and software engineering, with a focus on production-ready pipelines, scalable architectures, and tools that make data easier to explore and act on. A Streamlit Creator, Siavash has developed numerous Streamlit apps across both academia and industry, with several features in the Streamlit Gallery. He discovered Streamlit in 2019 at PyData and has been an early power user ever since. He is also deeply committed to knowledge-sharing, regularly writing technical articles on advanced Python techniques and data science best practices, while enjoying opportunities to teach, mentor, and help other data scientists and developers improve their craft. When he is not building things, he enjoys reading a good book, making coffee for friends, or getting nostalgic playing point-and-click adventure games.

Yuichiro Tachibana, known online as @whitphx, is a Developer Advocate at Hugging Face and an active open-source software contributor. His projects span areas such as web development, browser technologies, and developer tools. In the Streamlit community, Yuichiro created streamlit-WebRTC, a library that adds real-time video and audio processing features to Streamlit applications. Later, he developed Stlite, which enables Streamlit apps to run entirely in the browser through WebAssembly. At Hugging Face, Yuichiro applied knowledge gained from Stlite to create Gradio-Lite, which is essentially to Gradio what Stlite is to Streamlit. He has also developed Awesome Emacs Keymap, an extension that brings Emacs-style key bindings to Visual Studio Code. While different in focus from his Streamlit and Gradio-related work, it reflects his interest in improving developer experience across tools. Yuichiro regularly participates in the global Python community, attending and speaking at events such as PyCon. He values these opportunities to exchange ideas, and stay connected with developers from around the world. Through these projects and activities, Yuichiro has focused on making technical tools easier to use and more accessible, while contributing to communities that support open collaboration.

Souvik Roy has been deeply engaged with Machine Learning (ML) since the beginning of his career. He started out as an ML Intern at The Sparks Foundation, later contributing as a Streamlit Community Developer, where he focused on creating and optimizing automated ML models using Streamlit. Souvik completed his studies at the RCC Institute of Information Technology in 2024, before joining PwC as a Cybersecurity Engineer.

Over the course of his career, he has continued to explore the intersection of security and Artificial Intelligence (AI). Currently, Souvik is dedicated to developing ML-based SecOps solutions, and has also worked on deep learning–driven models, many of which are available on his GitHub profile. He is passionate about advancing secure, intelligent systems and applying his expertise and experience to drive innovation in cybersecurity through machine learning.

Khuyen Tran is a data scientist, educator, and author focused on helping teams build clean, reproducible, and production-ready data workflows. She has worked across data engineering, Machine Learning (ML), and developer advocacy at both startups and established companies, improving documentation, tooling, and educational content for modern data and forecasting workflows. 

Khuyen is passionate about turning complex data and ML concepts into simple, accessible, real-world solutions that teams can apply immediately. She is the creator of Code Cut and the author of Production-Ready Data Science, with more than 250 technical articles that have guided thousands of practitioners. Khuyen brings deep expertise in Python tooling, workflow design, and clear technical communication to the projects she supports.