Machine learning vision has been applied in different disciplines to automate sampling activities and agriculture production. This technology can automatically monitor and count different insects, reducing monitoring time and improving management practices. Different models can be applied to detect, identify, or count pests and beneficial insects, increasing precision agriculture use in Entomology. We developed frameworks with models and user-friendly web applications to detect and count insects in different crops that can be integrated into robotics for site-specific management, potentially revolutionizing traditional pest monitoring and management in field crops.