Wayve and Microsoft for Startups: Autonomous Driving
Authors: Aditya Sharma and John Stenlake
Last week, we announced that we are teaming up with the innovative London-based startup Wayve as they build their unique autonomous driving platform on Microsoft Azure. We have been working closely with Wayve for over a year now, going back to when they first joined Microsoft for Startups. It has been amazing to see Wayve grow their small but talented and passionate team during this time, as they have continued to impress the industry with one innovation after another.
Wayve first caught our attention back in 2018 when they released a short video demonstrating how they were able to use deep reinforcement learning (RL) to teach a car to follow lanes in under 20 minutes, using a single monocular camera as input, without relying on maps. This was the first time someone had successfully demonstrated how deep RL could be used to train autonomous vehicles. They may have been working out of a house in Cambridge with a small team, but we knew back then that Wayve was special. At a time when it seemed there were new autonomous driving (AD) companies starting up every month, Wayve stood out.
As we worked with Wayve, we focused on ensuring they had access to the best resources Microsoft had to offer to help them scale. During this time, Wayve moved their operations to London and are now driving across the UK with a fleet of autonomous vehicles training on petabytes of data. Given our strategic focus on the automotive industry and autonomous driving, we were able to share learnings from our other engagements across startups, OEMs and tier 1 suppliers with Wayve. We also learned a great deal from Wayve, and this engagement was influential in helping us design the Microsoft for Startups: Autonomous Driving initiative. When we launched the program at the Frankfurt Motorshow (IAA) last year, we brought Wayve over as one of its first members.
Taking a multi-faceted approach
An important aspect of autonomous development is the ability to scale your infrastructure. Broadly speaking, more available data leads to better performing algorithms and access to large amounts of compute enables you to train and deploy your models faster. There is a lot more to it however, and this is where Microsoft Azure becomes critical. The Wayve team wrote the following on their blog:
“It was clear across multiple facets such as cost, technology, and strategic alignment, that Microsoft Azure was the best choice for Wayve.”
Let us look at each of these factors in turn.
Cost: Cloud infrastructure is a big expense for AD companies. It is vitally important to make sure your cloud platform provides you great cost-saving options. The Wayve team specifically listed Azure Storage Reservations and Locally Redundant Storage (LRS) as examples of these. Price competitiveness is key, but it is even more important to have a flexible pricing model that allows you freedom of experimentation in the future. This is the core concept our offers are built around. Additionally, it is critical to ensure all your cloud workloads are optimized for cost performance. This is best achieved by carefully designing your architecture and working closely with our cloud architects and engineers. Programs like FastTrack for Azure are very helpful for this, especially when working on accelerated timelines.
Technology: AD startups are at the forefront of pushing the boundaries of cloud infrastructure requirements. You want to make sure your cloud provider always stays a few steps ahead of you on these. Azure achieves this in two important ways:
- We are committed to innovation and to supporting the latest and greatest to ensure our customers and partners have the bleeding edge of technology in their cloud offering. The Wayve team listed Graphcore's IPUs and NVIDIA's A100 GPUs as examples of this.
- As an AD startup partner, you get direct access to Microsoft engineers and program managers working on the Azure services you use. This not only provides you deep insight into how our services are designed and an opportunity to provide direct product feedback, but it also gives you a view of our roadmap and the ability to directly influence it.
Strategic alignment: Microsoft differentiates persistently in our focus on specific industries and our willingness to work very closely with customers and partners. Our ability to demonstrate our extensive portfolio of engagement and co-innovation in the automotive industry, our strategic efforts to ensure that customers are in control of their brand, IP, and data, our commitment to not compete with them in the future, and our track record as a partner-focused organization were material factors behind Wayve’s selection of Microsoft. All of this set the right tone for partnering with Wayve, as opposed to just selling them infrastructure.
The Microsoft for Startups: Autonomous Driving program puts together everything outlined above and more for our AD startup partners. You get direct access to Microsoft engineers and program managers, support from our cloud architects, opportunities to build solutions together, business enablement through access to our Automotive ecosystem, joint customer opportunities, marketing support, event showcases and more. Additionally, you get all the premium benefits from the standard Microsoft for Startups offering. Our startup partners are doing some amazing work, our goal is to provide them with all the enablement we can to help them be successful.
Read more about Wayve’s journey to scale using Microsoft Azure on the Wayve blog and hear directly from Alex Kendall, Wayve's CEO in our Transform blog. You can also find our live chat with Alex at Microsoft Ignite here. Visit Microsoft for Startups and Microsoft for Startups: Autonomous Driving for more information on our startup programs.