Neptyne builds a Python-powered spreadsheet for data scientists

Douwe Osinga and Jack Amadeo were working at Sidewalk Labs, Alphabet’s technology-advanced city-building company, when they came to the conclusion that most spreadsheet software wasn’t up to today’s data challenges. There are data science tools like Pandas and Jupyter Notebooks, but they tend to be too inaccessible for the layperson—at least in Osinga and Amadeo’s experience.

“Talk to any analyst or financial modeler and they’ll tell you that Excel just doesn’t work anymore,” Osinga said. “Everyone is aware of the need to move to more powerful solutions and Python is an obvious candidate. However, cooperation with today’s tools is insufficient.”

Osinga and Amadeo’s solution was Neptyne, an app that uses an AI assistant to help users program spreadsheets without learning to code. A member of the Y Combinator Winter 2023 class, Neptyne this month closed a $2 million pre-seed round from Y Combinator and a group of prominent angels, including Google AI head Jeff Dean and Google Maps co-founder Lars Rasmussen.

Neptyne joins a group of startups on a mission to reinvent the traditional spreadsheet. There’s Airtable, of course, and upstarts like, Actiondesk, and Pigment — the latter of which raised $73 million last November for its data analysis and visualization service. Recently, Equals, a San Francisco-based company, raised $16 million for its spreadsheet platform, which includes tools such as live data integration.


Image Credits: Neptune

Neptyne is different in that it contains a Python-based spreadsheet engine, Python being a programming language widely used for data science. Osinga describes it this way: It’s a spreadsheet where everything works the way you’d expect it to work in a spreadsheet, but it also offers access to the Python ecosystem—including libraries, frameworks, and tools.

“Large language models [like OpenAI’s GPT-4] who recently produced stunning results are very good at writing Python code,” Osinga said. “Because Neptyne is natively Python, it means that AI doesn’t just help you write formulas or visualize data—with AI, you can have a spreadsheet application dialog in front of you and change it for you.”

For example, imagine you have a pivot table built using Pandas, an open source Python data analysis library that summarizes data by product and region. With Neptyne, you can ask the AI ​​assistant to change the pivot table calculation, to include or exclude products, or to change the grouping criteria, for example by product category instead of by individual product. Neptyne updates calculations in real time, allowing you to explore different options.

“A lot of important modeling and calculations today are hidden in complex spreadsheets or data processing systems, and changing them requires experts to dig deep,” Osinga said. “Neptyne’s powerful AI integrations remove the limitations of conventional spreadsheets while dramatically reducing the complexity of using advanced data tools.”

These are fighting words. But it’s early days for Neptyne — the company is pre-revenue and doesn’t disclose the size of its customer base. Osinga is optimistic about growth this year, expecting to hire six employees to round out Neptyne’s current workforce of four.

Leave a Comment