Python in the Browser: Building Zero-Backend Data Apps with Pyodide

[prudhvi1709]

Abstract

Modern Python applications for data analysis and experimentation usually depend on servers — even when the workflows are exploratory, lightweight, or internal. This adds friction in the form of backend setup, deployment, and infrastructure management, especially when the goal is fast iteration and easy sharing.

This session explores how Pyodide enables running real Python code directly in the browser using WebAssembly, removing the need for backend services for an entire class of data and analytics workflows.

We will walk through practical patterns for building zero-backend Python applications, including:

Running CPython in the browser using Pyodide and understanding its constraints
Designing Python-first architectures where JavaScript acts only as orchestration
Managing application state, data ingestion, and computation fully client-side
Packaging and refactoring Python logic to work reliably in a browser runtime

The session will include a live demonstration using HypoForge, an interactive hypothesis generation and statistical testing tool, running entirely in the browser. The demo will show how Python can handle data ingestion, hypothesis generation, statistical testing, and result exploration without relying on servers, APIs, or cloud infrastructure.

By the end of the session, participants will understand how to:

Decide when Pyodide is a good fit for Python applications
Architect Python logic for browser-based execution
Avoid common performance and state-management pitfalls
Build lightweight, shareable data tools without backend dependencies

Category of the Talk

Developer Tools & Automation
Data Science, Machine Learning, and AI

Duration (Including Q&A)

25–30 minutes

Level of Audience

Intermediate (basic knowledge of Python and data workflows helps)

Speaker Bio

Prudhvi Krovvidi is a Data Scientist at Gramener, working on Python-based data and AI systems. He builds practical tools focused on automating data reasoning, improving developer productivity, and enabling faster experimentation using Python and LLMs.

He has developed and presented open-source projects such as SchemaForge and HypoForge, which explore schema reasoning, data quality automation, and interactive data workflows. Some of these experiments leverage Pyodide to run Python directly in the browser, making them lightweight and easy to try without backend setup.

Outside of work, he enjoys experimenting with emerging technologies and going on long bike rides.

Company: Gramener, Hyderabad
GitHub: https://github.com/prudhvi1709
Email: kprudhvi71@gmail.com
LinkedIn: Prudhvi Krovvidi
Experience: 1+ years

Prerequisites

Interest in Python-based tooling and data workflows
Basic familiarity with Python and data analysis concepts