Pydantic, the popular Python library and open source data-validation framework used by some of the world’s biggest companies, has a new commercial namesake and the backing of one of Silicon Valley’s most storied venture capital (VC) firms.
Pydantic Services Inc. emerges from stealth today with $4.7 million in seed funding led by Sequoia, with participation from Partech, Irregular Expressions and a host of angel investors, including Zapier co-founder Bryan Helmig, dbt Labs founder Tristan Handy and Sentry co-founder David Cramer.
London-based software developer Samuel Colvin started Pydantic as an experiment back in 2017, and in the intervening years the project has gone from strength to strength, used by developers at major tech companies including Alphabet, Amazon, Apple, Meta and Microsoft.
Its adoption has been driven in large part by FastAPI, a web framework for building APIs that integrates with Pydantic under the hood. However, Pydantic’s growth more broadly can also be attributed to the explosion of Python, which overtook Java in 2019 to become the second most popular programming language after JavaScript.
According to Colvin, Pydantic now garners some 48 million downloads each month and is used by 19 of the top 25 Nasdaq-listed companies.
“Right now, 12% of professional web developers use Pydantic across a wide range of applications,” Colvin explained to TechCrunch. “The pace at which developers have come to use and trust the tool showed me the scale of the problem and appetite for a better solution.”
So what, exactly, do developers use Pydantic for?
Well, there are myriad scenarios where online applications need to check and validate the type of data a user enters. For example, a simple online form that requests a name, email address and phone number will need to reduce the chances of someone entering the wrong kind of data, so it can be helpful if the form can automatically check the email is in a valid format, or the name-field is not left blank. Similarly, a bank might build a new system for processing transfers that collects data from multiple internal and external sources — this system has to ensure that the data is in the right format before any money transfer is executed.
To do this, Pydantic enforces Python’s “type hints” at runtime, which validates the data and serves up user-friendly error messages when an input is invalid.
“Pydantic allows developers to process external, untrusted data, making sure it conforms to an expected schema, and if it doesn’t, raises a helpful error,” Colvin said. “In essence, Pydantic makes working with real-world data much easier, and therefore faster — this saves many hours of work and avoids errors.”
‘Inspired by Pydantic’
Pydantic’s new commercial entity will incorporate a swath of new tools and services that are both “powered-by and inspired-by the Pydantic library,” according to Colvin, who said that he expects the first fruits of this labor to be made available later this year.
“We’re building cloud services, and we’ll have a generous free tier and usage-based pricing after that,” Colvin continued. “We’ll make developing and deploying applications to the cloud easier, safer, faster and ultimately more enjoyable for developers. We’ll start by helping engineers with small applications or functions, but long-term our aim is to be a force-multiplier for all developers — giving them the tools that allow them to improve the world for everyone.”
So, what we’re probably talking about here, in the longer term at least, is something akin to a platform-as-a-service (PaaS) along the similar kind of lines as Salesforce-owned Heroku.
Colvin has already been working on Pydantic full-time since last March, funded through a combination of savings and corporate sponsorships, including cash infusions from industry heavyweights such as GitHub (Microsoft), AWS and Salesforce.
On top of that, the open source project has garnered significant code contributions from more than 351 separate entities, including developers at Google, AWS, Visa, and Stripe. This positions Pydantic strongly as it looks to build out a full-time team — any open source project that has such industrial gravitas usually stands a good chance of attracting top technical talent.
“Pydantic’s contributors would be the envy of any major tech company, and my first few hires will all be developers who’ve made significant contributions to the project,” Colvin said. “In effect, the network and reputation of Pydantic allows me to hire engineers who would otherwise only be available to companies with the biggest names and deepest pockets.”
Pydantic will start with an initial team of six, with the first three engineers based in Montana, Chicago and Berlin.
“I’m hiring the best developers I’ve met in open source, hence they’re all over the world,” Colvin noted.
Show me the money
Securing the backing of one of Silicon Valley’s most illustrious VCs is a major coup for any fledgling startup. Indeed, Sequoia has previously backed the likes of Apple, Google, Cisco, Dropbox, Electronic Arts, PayPal, Zoom and WhatsApp, while in recent years it has been doubling down on its European efforts with new region-specific partners.
Today, Sequoia has five partners based out of its Marylebone office, however, its investment in Pydantic was led by U.S. partner Bogomil Balkansky, who was keen to highlight Sequoia’s history of investing in startups with open source foundations, including MongoDB, Confluent and dbt Labs (formerly Fishtown Analytics).
“Sequoia has been thinking about the ‘rise of the developer’ for more than a decade, and we have partnered with many open source-based companies,” Balkansky said in a statement issued to TechCrunch. “We are thrilled to partner with Samuel because of this amazing track record creating the widely used and beloved Python data validation library Pydantic.”
Today’s news comes just a few weeks after Sequoia announced a $195 million fund dedicated to seed-stage startups in the U.S. and Europe. Its fifth seed fund, Sequoia also said that the money would help fund startups in its Arc program, a London and Silicon Valley-based program it launched last year to discover and mentor so-called “outlier” startups across the U.S. and Europe.
However, Sequoia didn’t confirm whether its investment in Pydantic stems from that new fund.
It’s worth noting here that although Sequoia has been looking to invest in European founders, the new Pydantic Services Inc. entity will be incorporated in the U.S., though Colvin will remain in the U.K. for the time-being.
“A number of the early employees are based in the U.S., and it’s easier to give them share options if it’s a U.S. company,” Colvin said. “If the company is successful, it’s likely we’d need to move it to the U.S. in the future, [and] I’m told that’s complex and expensive, so it seemed sensible to start off with a U.S.-based company.”
With $4.7 million in the bank, Colvin said that they’re continuing to rewrite parts of Pydantic in Rust, with a view toward making it more efficient via a ten-fold performance improvement. So while Pydantic 2.0, which is scheduled to be released later this year, will still be a library for Python developers, some of its core logic will be written in Rust.
“Making Pydantic faster should significantly reduce the amount of energy consumed by the servers running applications which are built on Pydantic,” Colvin said. “I’m a strong believer that Python is a great language for application development, but as library developers, we can significantly improve those libraries — make them faster, safer and less energy-intensive to run — by building tools and services for those applications using fast and secure languages like Rust.”