Aetina introduced its end-to-end AI management solution in an NVIDIA GTC on-demand session this week. The solution helps global AI partners and clients of Aetina successfully adopt edge AI using NVIDIA AI development and deployment tools, as well as Aetina's NVIDIA AI-powered training and inference platforms. To adopt edge AI, system integrators and developers need to train AI models and deploy them on edge devices. The AI model training process involves collecting and labeling large amounts of data using high-performance computing platforms, which can result in high training costs.

These factors make the training process challenging and time consuming for system integrators and developers. AI model deployment can also be difficult when the system integrators and developers have multiple remote edge devices in different locations. The challenges of AI model deployment include optimizing models for inference to run efficiently on edge devices, long deployment processes, difficulties of team communication and collaboration, high follow-up monitoring costs caused by large number of AI models, security issues, and high costs.

End-to-End AI Management Solution Aetina provides an end-to-end solution to help its partners and clients develop and deploy AI applications for the edge. The solution consists of Aetina's NVIDIA-Certified edge computing platforms and NVIDIA's AI model development and deployment tools. These tools include NVIDIA Fleet Command™, and the NVIDIA AI Enterprise software suite, which provides enterprise support for the NVIDIA TAO toolkit and NVIDIA Triton Inference Server™.

With Aetina's NVIDIA-Certified edge computing platforms and other NVIDIA tools, Aetina streamlines the AI training and deployment process, helping partners and clients adopt edge AI more quickly. An Aetina solution team first trained an AI model capable of finding defective products in factory production lines for the AOI system with NVIDIA TAO on Aetina's AI training platform—SuperEdge AIP-D422—and then uploaded the model to NVIDIA GPU Cloud (NGC). With NVIDIA Fleet Command™, the solution team remotely deployed the model on Aetina's AI inference platform—MegaEdge AIP-FQ47—in the factory of Aetina's client from NGC, successfully developing the prototype of the AOI system.

With AOI cameras, the prototype of the system is now able to greatly speed up the inspection task of finding the DRAM products with incorrectly placed corporate stickers with high accuracy in less than one second, compared to a human inspector that usually takes about ten seconds to do the same task. When the AI-powered AOI system is fully built in the future, it will be installed in the factory of Aetina's client to run inspection tasks in multiple production lines. In addition, NVIDIA Fleet Command™ allows the Aetina solution team to quickly update the AI model on edge devices in the factory production lines of Aetina's client, enabling efficient remote AI management—a key to successful edge AI deployment.

The end-to-end AI management solution is a part of Aetina Pro-AI Service—which helps global partners and clients adopt AI for different vertical applications besides AOI in factories, with Aetina's edge AI hardware and software.