Argo AI Case Study

About Argo AI

Argo AI is a self-driving technology company on a mission to design and build a safer, more
accessible, affordable and convenient way to get around. With the goal to scale quickly, Argo AI
is developing its software to be reliable and robust from the start. As early as 2021, vehicles with
fully integrated Argo AI technology will be operating in multiple cities as a platform for ride hailing
and goods delivery services.

The Challenge

In order to develop robust software for large-scale commercial operations, Argo AI needs to
frequently test and analyze pull requests, and create application programs for new software
releases. In order to deliver high-quality code more quickly, the team sought a solution to speed
up their builds, tests, and QA script testing. With development across multiple OS platforms, a
compatible acceleration solution was needed.

How IncrediBuild crunched it

By using IncrediBuild to distribute computing tech on the cloud, Argo AI’s infrastructure team was
able to get on-demand CPU boosts from thousands of cores. In addition, the team was able to
harness their on-premise idle machine cores for more CPU power. This proved to be an adaptive,
cost-effective way to get speed on demand.

“It was very easy to leverage the cloud with IncrediBuild. We are able to scale up cloud instance
CPU cores on the IncrediBuild network when our load gets high,” says David Schwanky, who
leads Argo AI’s software infrastructure team. “This helps us keep our cloud costs in check, while
still providing fast builds.”

As IncrediBuild users, the team’s build time has improved eightfold.
“IncrediBuild has been critical in speeding up the entire continuous integration and test cycle,
and has enabled our developers to land more code in a shorter amount of time,” says Schwanky.
“We use IncrediBuild across our development team for local builds, cloud CI builds, and running
tests. We were able to easily integrate IncrediBuild into our existing build system with minimal

With their development streamlined, the team has been able to implement a more efficient code
analysis process, and eliminate bottlenecks in their build pipeline.

According to Schwanky, “Using IncrediBuild has allowed us to speed up our build system to the
point where we are able to run more testing and analysis on each pull request. This allows us to
deliver higher quality code much faster.”

And when it came to choosing a Linux tailored solution, IncrediBuild was the right fit. “We use
Linux for the depth and breadth of open source software it provides.”

The Bottom Line

Since switching to IncrediBuild, Argo AI has been able to meet the demands of a rapidly growing
company. By scaling into the cloud with IncrediBuild and utilizing hundreds of cores, Argo AI’s
infrastructure team is able to develop and test software for autonomous vehicles without having
to choose between speed and qualhigh-qualitydiBuild allows us to scale our build system and deliver high quality code, faster.”

After Incredibuild
Build time: 8x Faster
Test time: 4x Faster