Yandex and five national research universities — HSE, MAI, MEPhI, MIPT and ITMO — we have developed a model of competencies in the field of physical AI. This is the first in Here is a complete description of what the creators of robots and autonomous systems should know and be able to do. The document is open to everyone: any university can use it to create educational programs. The first program based on the model was Yandex Physical AI Garage. Recruitment for it has already started and will last until the end of summer.
The model describes how roles are distributed in teams dealing with physical artificial intelligence, and what fundamental knowledge and applied skills are required in each role, from engineer to manager. For example, an ML researcher should not only master linear algebra and understand neural network architectures, but also know kinematics, understand how sensors work, and be able to work with popular robotics frameworks.
Thanks to the model, universities and industry will have a common coordinate system. Universities will be able to update their curricula in order to train specialists who are in demand on the market. Companies developing physical AI will have the opportunity to fill the personnel gap and reduce the time for young professionals to immerse themselves in the specifics of the profession.
Partner universities are already using the model. Other educational institutions may also consult it when designing new or updating existing programs. For example, a university that trains backend developers will be able to launch an additional track specializing in physical AI. For example, it will teach students how to create APIs for the interaction of ML models and robotic platforms.
Daria Kozlova,
Director of Yandex Education
AI is increasingly penetrating the physical world and finding applications in manufacturing, transportation, medicine, and beyond. In this regard, there is a growing need for people who can work with both algorithms and sensors, and know how to implement AI safely. Our model combines the experience gained in creating physical AI in Yandex, with a fundamental education from leading universities, and allows you to graduate specialists who are ready for the challenges of the industry.
The model formed the basis of the Yandex Physical AI Garage program, which Yandex launched together with partner universities. The program combines fundamental courses that will be taught at universities, and practical modules from experts from Yandex. Its participants will work on real—world tasks, such as improving perception and navigation models that help autonomous vehicles navigate the terrain. Students from the specialized fields of HSE, MAI, MEPhI, MIPT and ITMO will be able to participate in the Yandex Physical AI Garage. The program provides 100 places, and participants receive an additional scholarship.
Yandex Education systematically develops projects for teaching physical AI. AI robotics clubs have been opened for schoolchildren in different regions of Russia and an online platform with a robot simulator has been launched. Students can enroll in the Master's degree in AI Robotics at Skoltech, created with the participation of Yandex, and take part in student camps. Yandex Education, together with 13 universities, is also creating modules on physical AI. In the new academic year, more than a thousand students will begin to master them.
Please note that this press release is based on materials provided by the company. AK&M Information Agency shall not be held liable for its contents, nor for the legal and other consequences of its publication.
