Watch’s self-driving car handle California city streets on a rainy night

Autonomous vehicle system startup is ready to show off its technology in use for the first time, with a video showing the company’s test vehicle making its way through the streets of Mountain View.

Other companies have shown us similar videos, of course, including GM-owned Cruise very recently, and Tesla. But’s demonstration is interesting because it depicts incident-free, unbroken footage shot at night, including extensive segments where it’s raining quite a bit.

These are impressive because inclement weather and nighttime driving are particularly challenging for autonomous driving software, owing to factors like reduced visibility for optical sensors and a lot more interference with sensors in general because of the added noise created by the rain. Road conditions are also obviously affected by precipitation, and requires a different style of driving to navigate effectively (something a lot of human California drivers seem surprised to learn).

Some of the majorly impressive moments in the video include a point where another vehicle cuts in front of the car at a four-way stop intersection, and another near the two-minute mark where the car successfully manages a red light that’s knocked out and acting as a four-way stop as a result.

In case you missed’s launch in August last year, the company is employing deep learning and neural networks to teach its autonomous driving systems, and is focused on using said technologies not only for identifying and detecting objects, but also for decision-making and even communication with other drivers and passengers, an area it thinks is just as important as the driving itself. CEO Sameep Tandon provided the following statement regarding the video and the startup’s testing to TechCrunch:

Today we released the first public video of self-driving technology from the team. In the video, you will see our system navigate the streets of Mountain View, CA not only in daylight, but also in the rain and at night. These are conditions drivers encounter every single day. Any successful self-driving technology will need to address countless unpredictable situations and a wide range of driving conditions, yet few are able to today.’s unique deep learning technology processes information and makes decisions more like the human brain does. Our technology allows us to scale to all kinds of challenging environments.

As one of the few startups in California with an Autonomous Vehicle Testing Permit, we are prioritizing our responsibility to develop technology that is safe in all conditions. We know that our team has the expertise to create the future of autonomous transportation. We couldn’t be more excited to share this first look – and bring you along for the ride!