A variety of new career opportunities are emerging amid the development of China's artificial intelligence (AI) industry, from training robotic arms to building virtual training areas for inspection robots.
Hot Topics
In September 2026, China's Ministry of Human Resources and Social Security and other government agencies announced 11 new professions, including tangible AI robot application technicians and digital twin engineers.
At Sevnce Robotics, which operates in Liangjiang New Area, Chongqing City, southwest China, Tang Zixiang works as a tangible AI robot application engineer, training tangible AI robots to perform various tasks.
Tang will adjust the position and angle of several knobs to simulate various working conditions and use the resulting data to train AI models. It then applies these trained models to robotic arms for further testing and refinement.
Through a series of repeated tests, the robots slowly learn tasks in simulation before applying them in the real world.
"We're tasked with helping robots move from the test room to broader applications in the real world," Tang said.
Sevnce Robotics develops specialist robots for the oil, gas, chemical, and energy industries. Their machines can replace or assist human supervisors in hazardous environments.
In such environments, robots must be able to operate reliably over long periods. Tang and his colleagues needed to set speed and accuracy requirements, test whether the robots could recover from operational failures, and determine when to stop and/or require human intervention.
In addition to tangible AI robot implementation engineers, there are also digital twin engineers, another newly recognized profession that is generally tasked with helping prepare robots for real-world deployment.
Jiang Lei, who specializes in this field, is tasked with building a virtual environment where inspection robots can retrain their tasks before deployment.
Using three-dimensional (3D) reconstruction, Jiang's team built a digital mock-up of the factory, allowing the robots to rehearse their inspection task routes and check whether their cameras could capture instrument readings. Once these tasks were validated in the simulation, they could be transferred to the physical robots.
Typically, technicians will bring the robot to a job site before mapping the surrounding environment, plotting a route and determining checkpoints, Jiang said.
However, digital twins allow these tasks to be retrained through simulation, reducing the time from 2-3 weeks to just 2-3 days. The simulations also generate data for testing and training algorithms, reducing the time and cost of data collection in real-world environments.
As the industrial chain becomes more specialized, career opportunities are opening up for people from a variety of educational backgrounds.
Jiang, for example, studied vehicle mechanical engineering at university and moved from autonomous driving research to robot development, which requires expertise in 3D modeling, programming, and simulation.
Meanwhile, Tang, who has a bachelor's and master's degree in computer science, applies reinforcement learning and imitation to robot training while continuing to explore robot kinematics and motion control.
The growth of the tangible AI industry also increases the need for experts in data collection, model training, and system operation, maintenance, and setup.
China's Ministry of Industry and Information Technology recently released data showing that there are more than 400 models of humanoid robots in the country and throughout the first half of 2026, 40,000 humanoid robots have been produced.
Tags:
technology