Discover more from Data Engineering Weekly
The Data Founder Story: From McLaren Formula 1 to Quix
By Tomas Neubauer
Hey there, I’m Tomas, one of the Quix founders.
After several years of hard work, Quix has become a data processing platform that allows engineers and scientists to transform and deliver data as it’s created. It’s designed around a message broker, differentiating it from traditional data engineering platforms designed around databases. The platform is commonly used to automate digital experiences that produce a large volume of real-time data and works best when aggregations aren’t enough.
Here are how those years of hard work went, taking my co-founders and me from a high-speed racing company that used data to build a data company of our own.
Racing to act on data as it’s created
Around 2018, Peter Nagy, Patrick Mira Pedrol, Mike Rosam and I began building systems that most Formula 1 teams now use to stream and process over a million data points per second in real-time. Formula 1, where a team’s ability to connect engineers to streaming data directly correlates to their performance on the track, is the ultimate live-data environment.
At McLaren, we also consulted on projects to apply real-time ML in the sport, automotive, transportation and healthcare industries. We helped clients use real-time data to improve fan engagement, develop electric vehicles faster, predict tire failures in vehicle fleets, and optimize patient postoperative outcomes.
Across these experiences, we kept seeing the same pattern: the rapid adoption of message broker technologies increasing the value of data-driven operations in every industry, together with the explosion of people with Python skills, the language of machine learning. The two trends should be a match made in heaven, but there was a problem.
The problem: Complicated streaming solutions built from scratch require too much time and money
Machine learning on streaming data in real-time is an order of magnitude more complex than ML on batch data. And while our data scientists could quickly develop ML models off-line, it took us years to operationalize machine learning on streaming data infrastructure.
We’ve since met with many companies working on real-time machine learning projects, and they all have similar problems. ML-first companies like Uber and Airbnb invest hundreds of millions to develop bespoke internal solutions, while huge legacy organizations in every industry pay suppliers eye-watering sums for never-ending digital transformation projects. And while broker technology is becoming more accessible, operationalizing real-time ML on streaming data remains entirely out of reach to all but the top 1% of ML-first companies. We founded Quix to solve this problem by helping any Python developer work with streaming data. Quix is a developer-first platform with the message broker at the core. It accelerates the development of real-time data-driven products by providing all the infrastructure, APIs and SDK that Python developers need to stream, process and store data without support from any IT, DataOps or DevOps people.
Our solution? A single platform for all
In 2020, in the depth of the first lockdowns, we went out on our own to build a generic platform that anyone developing streaming data apps could use. We wanted to decrease the amount of time and money companies poured out to harness the power of stream processing while enabling individuals to access the same technological benefits.
We’re focused on removing the time, hassle and investment of setting up infrastructure and integrating different technologies so that any organization — regardless of size, funding or industry — has the ability to process stream data in real-time with machine learning models and automate applications with the same quality and reliability as the world’s leading ML companies.
All rights reserved Pixel Impex Inc, India. Links are provided for informational purposes and do not imply endorsement. The data founder story contents provided by the featuring data founders in this article and Data Engineering Weekly is not responsible for any compliance with applicable laws, rules, and regulations.