Helm.ai

Inaugural AutoTech Breakthrough Awards program recognizes Helm.ai for Deep Teaching

Image for post
Image for post

Some exciting news!

Today we’re honored to announce Helm.ai has been named the ‘Autonomous Driving Solution of the Year’ by the inaugural AutoTech Breakthrough Awards program conducted by the Tech Breakthrough organization, a leading market intelligence and recognition platform for the most innovative technology companies in the world.

“Our Deep Teaching technology addresses core difficulties impacting the autonomous driving industry, quickly resolving corner cases at the tail end and accelerating the roadmap to advanced L2/L3 production systems and L4 full autonomy.” said Vladislav Voroninski, CEO, Helm.ai. “Using our approach, we have topped out public computer vision benchmarks and surpassed state of the art production systems with minimal engineering effort. …


Get caught up on Helm.ai’s latest developments

Image for post
Image for post
Photo by CHUTTERSNAP on Unsplash

It goes without saying, but it’s been a busy year for us here at Helm.ai.

We emerged from stealth in late March, with TechCrunch breaking the news of our $13M seed round with investments from various notable venture capital firms and investors, followed shortly thereafter by the announcement of Deep Teaching, our breakthrough approach to Unsupervised Learning and AI.

We joined other visionaries to discuss our perspective on Unsupervised Learning during The Information’s 4th Annual AV Summit, hosted a Reddit AMA with Helm.ai …


Deep Teaching’s Potential for Automation Extends Beyond Automotive

Image for post
Image for post

Helm.ai’s Deep Teaching methodology represents a fundamentally different approach to AI which extends well beyond its core use today in self-driving vehicles.

Helm.ai’s scalable approach to building AI software allows training neural networks on huge datasets without annotation or simulation, resulting in higher levels of robustness and generalization than achievable with traditional approaches. Moreover, Deep Teaching can be applied to sensor data of any environment, including applications beyond land-based navigation. Described by Forbes as the ‘Android Of Self-Driving Cars’, Helm.ai is building an AI for multiple trillion-dollar markets that is rooted in unsupervised learning, which is 100,000x more capital-efficient than traditional AI approaches. Applications of Deep Teaching range from autonomous driving, to drones and consumer robots and even retail checkout. …


Join us as we discuss the future of mobility at the Drive Sweden Forum

Image for post
Image for post

The global mobility industry is due for an exciting, turbulent future.

People and goods are about to engage in a transformational shift in how we get from point A, to point B, and beyond. Strategic innovation will be mission critical, and our Deep Teaching unsupervised learning methodology is poised to power a wave of scalable autonomous vehicle technology that will shape the road ahead.

Mobility solution companies can expect, and should prepare for, substantial market shifts, regulatory uncertainty, breakthrough technological development and application, and significant changes in consumer behavior.

New opportunities will necessitate the products of tomorrow, and the products of tomorrow will generate new opportunities. Market demand is accelerating and shifting heavily in favor of more autonomous mobility solutions. It’s a thrilling time to be at the vanguard of what lies ahead, shifting the way we move through the world. …


We’re excited to announce we’re collaborating with Honda Xcelerator

Image for post
Image for post

One of the most exciting things about being a startup is bringing new ideas to big companies that value and foster innovation. That’s why we are excited about the work we are doing with Honda Xcelerator, Honda’s Silicon Valley-based program that serves as a sort of entry point to global Honda for startups and other innovators. The idea is if you’re an advanced startup like Helm.ai, Honda Xcelerator can help you quickly begin tangible collaborations. But like more traditional accelerators, if you’re a startup with an idea that needs more baking or backing, they might help you bake or back it. …


Our thoughts on fleet data, Supervised Learning and the adaptability of Deep Teaching

Image for post
Image for post

Autonomous driving companies relying on fleet data, human-annotated data and simulations to scale are doomed to fail.

Let’s unpack that.

For more than a decade autonomous driving companies have poured billions of dollars on assembling fleets of vehicles to better understand driving patterns, pedestrian behaviors and many other data points. Armies of human annotators have been recruited to manually annotate data and assess simulations for many more billions. …


As a member of the NVIDIA Inception Program, we’re reimagining the development of scalable AI

Image for post
Image for post

Our journey towards safe, interpretable, scalable and cost-effective autonomous driving technologies continues to evolve, and Unsupervised Learning is the future for self-driving AI software.

Unsupervised Learning allows us to train neural networks without labeled data. It’s an open area of AI research, with some previous approaches aiming to identify new patterns in data or pre-processing a pool of data.

Through Deep Teaching, our proprietary Unsupervised Learning method, it’s our aim to dramatically reduce time and cost bottlenecks in autonomous vehicle development cycles … chiefly, we’re lessening the burden on human annotators.

Our NVIDIA Partnership

As a member of the NVIDIA Inception Program, Helm.ai is reimagining the development of scalable AI software. In our data center, we use high-performance NVIDIA V100 Tensor Core GPUs to run the unsupervised learning techniques that train our self-driving algorithms. This helps make it possible to process petabytes of data without experiencing costly roadblocks. …


Here’s what we discussed when we met virtually with drivers of autonomous vehicle technology at a SEMICON West panel.

Image for post
Image for post

On July 22, we took the virtual stage at the 50th annual SEMICON West conference, which featured leading global microelectronics manufacturing and design professionals.

Our panel, in particular, consisted of Silicon Valley mobility startups and VCs, as we discussed our innovations, experiences, growth opportunities, and COVID-19 pandemic-related challenges.

Joining us on the panel were Junli Gu, Ph.D. (moderator), Jeff Peters, Ph.D. (Partner, AutoTech Ventures), Yao Zhai, Ph.D. (Cofounder, Ziiko Robotics), Bibhrajit Halder, Ph.D. (Cofounder & CEO, Safe.ai), Miao Hong, Ph.D. (General Partner and Chairman, Silicon Valley Future Capital), and Yvonne Lutsch (Investment Principal, Robert Bosch Venture Capital).

Part of our discourse involved discussing the different ways in which we approach developing the software required to make autonomous driving possible. We create full-stack software solutions that can also be dispensed modularly, starting at L2+ and aiming for L4 stacks that can scale beyond autonomous vehicles, including fields like drones and robotics. …


We were already forecasting, and shaping, major shifts in our industry this year. See how COVID-19 is accelerating that shift even more.

Image for post
Image for post

The coronavirus pandemic, in addition to taking its exceedingly tragic human toll, is forcing our industry to face even more major, novel challenges. According to a recent McKinsey report, mobility solution companies can expect, and should prepare for, market consolidation, regulatory uncertainty, suspensions in technology development and testing, and significant changes in consumer behavior. Much of these changes were already in play, yet the pandemic hastened the pace of this evolution.

While coronavirus presents new obstacles and impediments for other startups and even some major players, it also presents new opportunities. …


Annotation is an expensive bottleneck. There’s a workaround.

Image for post
Image for post

“How does your AI drive on a road it’s never seen before?”

That’s the first question posed to Helm.ai CEO Vlad Voroninski by TechFirst podcast host John Koetsier during Vlad’s recent guest appearance.

It’s a great question, of course, and offered us a unique springboard into talking about other opportunities for our proprietary Deep Teaching technologies for Unsupervised Learning.

The answer is almost better rephrased as a question: “Well, how do you drive on a road you’ve never seen before?”

You learned how to drive, but that’s not all. First, you learned how to see, and then how to recognize and interpret the things that appear in your field of vision, and then you learned how to make the mechanical movements to react to those objects. …

About

Helm.ai

Helm.ai is building the next generation of AI technology for automation.

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store