Jensen Huang, Nvidia's CEO, has introduced a new test for engineers at Google Cloud Next, a process that involves setting a baseline salary and then assessing an engineer's ability to consume technology resources. If an engineer is able to reduce their annual AI token consumption by more than 50% from the baseline amount, they may be deemed "worth keeping." The idea is to identify which engineers are efficient with their budgets, rather than just focusing on their technical skills.
The process involves setting a baseline salary for each employee and then tracking how much AI tokens they use per year. If an engineer's token consumption falls below 50% of their salary, it may indicate that they have been using too many resources to justify their compensation package. On the other hand, if an engineer is able to reduce their token usage by more than 50%, it could suggest that they are efficient with their budget.
This new test is part of a larger effort to improve efficiency in the tech industry. By focusing on reducing costs and optimizing resource utilization, companies can potentially increase productivity and competitiveness. The use of AI token consumption as a benchmark for employee value may also raise questions about the role of technology in determining job worthiness, but it seems that Huang is using this metric to identify high-value employees who are willing to make significant sacrifices in terms of their time and resources.