Contextualising Code with Step Size. An Interview with Alexandre Omeyer
Hi Alex, firstly could you tell us a little about you and your partners at Stepsize's background?
Sure. There are four co-founders at Stepsize; Nicolas Omeyer (in charge of Product), Matt (our CTO) and Jared (in charge of Front-end dev & Design), and myself (CEO).
Nick, Matt and Jared all studied Mathematics at UCL. Jared earned an MSc in Mathematics from UCL whereas Matt decided to move on to the Computational Statistics and Machine Learning (CSML) MSc at UCL after spending a year at Oxford to get his MSc in Financial Mathematics. Nick went straight from Mathematics to the CSML MSc at UCL. I studied Economics, Business, Management and English for my BSc at the University of Paris West before moving to the UK where I completed an MSc in Entrepreneurship and Management at the University of Sussex.
We moved in together as soon as we graduated and started working for various tech companies. Matt was LoopMe’s (a mobile advertising startup) first Data Scientist hire, Nick was working on the lending decision models at the business short term lender Everline, Jared was dealing with large scale software integrations for a procurement company (BravoSolution), and I was doing Business Development IOVOX and later DueDil.
Around that time, we also started working on all sorts of side projects together when we were home from work. We built things like a marketplace to match students with tutors, a news recommendation engine that used Twitter data to recommend the best content, and a few other prototypes for some of our ideas. Through that process, we all taught ourselves how to code and eventually stumbled across the concept that later became Stepsize.
What exactly does Stepsize do?
At Stepsize, we build tools for software developers to help them build better software faster.
Our first product, Layer, is a desktop application for devs that automatically adds context to codebases. It does this by hooking up with all the tools devs use to build software, structuring the historical data contained in these tools, and attaching it to the code.
As more users and customers trust us with their data, we’ll introduce smart features (that’s where the AI and ML come in) to simplify the entire development process. Think of tools to help you predict how long a given feature will take to build, or assign the right members of your dev team to a task, or automatically collect the right internal and external information they need to build and ship faster.
Where did the idea behind Stepsize come from, and what would you define at Stepsize's mission?
I’ll start by answering the second part of your question: Stepsize’s mission is to simplify software development to make it more efficient, powerful, and accessible. What this means is that we ultimately want to abstract away programming languages to make software development universally accessible.
To give you a bit more context, as I explained earlier, the four of us moved in together as soon as we graduated from university, and worked on a bunch of side projects after hours with the ambition of one day building a business. While taking an online course on deep learning for natural language processing, Nick started wondering about whether we could apply similar techniques to programming languages which have more structure and less ambiguity than English or French.
Being self-taught developers, we were intimately familiar with the huge learning curve of software development, and it’s such a key driver of progress that the idea of making it more accessible and efficient was really exciting. We started bouncing ideas on how to apply machine learning to the software development process and Stepsize was born.
After playing around with neural networks using publicly available data from Stack Overflow and GitHub, we concluded that the right approach was to gather and generate the right data that would allow us to iteratively streamline and automate various parts of the software development process. With this approach, programming languages would eventually be abstracted away completely and we’d build a very valuable and important business on the way there.
I’d like to stress how important this is. To us, the radical simplification of software development is one of the best things that could happen to the world. The products we’re building today are for developers, but think about the impact on society if we manage to simplify software development to the point where it’s accessible to anyone with access to a computer. We’re talking about millions or billions of people being able to create with software, not just the 20 million or so developers capable of writing code today.
So we set out to figure out how to help developers with their daily work while performing this data collection exercise. After a fair bit of exploration, a few prototypes and speaking to many developers working in teams, we eventually came to the realisation that every day they have to contribute to codebases they’re largely unfamiliar with, but no tool specialises in helping them understand the past – the who, why, and when of any given piece of code. We then started building what later became our first product, Stepsize Layer, to contextualise code and allow developers to work on shared codebases perfectly informed.
Stepsize has already attracted backing from a number of major players such as Virgin Media and Techstars. What do you think is the key to creating a valuable product and service like Stepsize?
I think this topic has been very well documented online and is somewhat unique to each situation, so I’ll give it a shot but forgive me if I simply end up repeating something you’ve read before.
In my opinion, there are several ingredients that allowed us at Stepsize to get to where we’re at.
Firstly, we allowed for ample exploration and experimentation. This is what we did together after university. We took the time to discover what we were truly passionate about, where we felt we could contribute most, and gain a deep understanding of the industry and people we were looking to serve. From this point, we were able to identify something that we thought we could significantly improve on and that we were so passionate about that it was worth committing our professional lives to. Getting this right from the get go means that the entire team is committed for the long run.
Secondly, don’t wait for anyone to get started. Nowadays, you don’t need to wait for investors to start a software company. If you don’t have a technical co-founder, learn to code and build your own prototype.
Stay curious, focused, and keep learning and you’ll be able to gradually forge what was only an idea into a valuable product.
What are your plans for Stepsize as we move into 2017 and beyond?
Our first goal is to release the Layer beta to the public. The product is currently in private beta, we’re getting close. As this version of the app will be available for free, we’ll remain focused on Layer and aim to deliver a fully featured version that we can charge for soon after.
This is where things will become even more interesting. In the medium term, our goal is to leverage the data our users and customers trusted us with to make Layer evolve into a tool to reduce the cognitive load on developers by making the information they need available when they need it (think Google Now for devs), as well as assisting them with their daily tasks. They will become much more efficient and effective.
As for the long term, Stepsize is working towards a future where software development is available as a service thanks to AI. Literal software as a service. Software development will look very different; code will be abstracted away and anyone should be able to collaborate with an intelligent agent to bring their idea to life. There’s obviously a very long way to go, but we’re convinced that we’re tackling the problem the right way and will get there eventually. Early signs are very promising.
To be featured or find out more:
e-mail us on email@example.com
call us on 0845 643 5375
or contact Janice on Linkedin
First published on Company Connecting December 2016