In order to control AI-related expenses, large enterprises are limiting employee usage quotas, choosing smaller models with lower costs, and using route schedulers to process only the most difficult query requests to the most expensive large models. Cognition, the startup behind Devin, the AI code intelligence agency, has launched a new solution. The company revealed that by measuring the efficiency of different engineers using AI, companies can allocate more resources to the most efficient employees. Cognition Chief Technology Officer Calvin Giroux said, “User usage limits should not be fixed, but should be adjusted dynamically based on usage efficiency and output value.”

In order to control AI-related expenses, large enterprises are limiting employee usage quotas, choosing smaller models with lower costs, and using route schedulers to process only the most difficult query requests to the most expensive large models. Cognition, the startup behind Devin, the AI code intelligence agency, has launched a new solution. The company revealed that by measuring the efficiency of different engineers using AI, companies can allocate more resources to the most efficient employees. Cognition Chief Technology Officer Calvin Giroux said, “User usage limits should not be fixed, but should be adjusted dynamically based on usage efficiency and output value.”