We envision that vision-language models as powerful as ChatGPT will soon be running in the edge, enabling autonomous cameras, drones, humanoid and sea robots deployed anywhere to reason locally while performing complex tasks. This new technology will also transform manufacturing, retail, and supply/chain.
We have the mission to deliver the power of foundational Vision AI models anywhere in the physical world
Nauman did a PhD in Electrical and Computer Engineering from Purdue University, with a focus on computer architecture and operating systems – he solved multicore CPU resource management problems, and proposed solution adopted by Intel, AMD, and Google.
After his PhD, he joined Google where he worked for 16 years. At Google, he played a key role in scaling of storage IO at Google by enabling high performance SSDs and distributed storage.
He also built Google baremetal cloud service from the ground up, to unblock multi-billion dollar deals for Google Cloud. He led all functions for this product including cross functional teams (sales, support, etc) and owned relationships with external vendors and enterprise customers
Leo is a passionate tech leader and hands-on scientist with over two decades of experience developing real-world AI systems and high-performance computing. He has built cutting-edge computer vision solutions, spanning robotics, autonomous systems, and cloud. Most of his career has been spent in the tech industry, where he has consistently translated complex research into scalable AI products, generating 8-figure revenues globally. Leo has been a founding member of multiple incubation teams at Microsoft and Amazon, taking AI initiatives from concept to launch. Earlier in his career, he contributed to the development of optimized numerical linear algebra libraries (BLAS) for the IBM Blue Gene/L supercomputer at the T.J. Watson Research Center. He earned his Ph.D. in Electrical and Computer Engineering from Purdue University.
Visum AI envisions a future where vision-language models (VLMs) as powerful as ChatGPT run directly on edge devices—enabling autonomous cameras, drones, humanoid robots, and sea robots to reason and make decisions locally while performing complex tasks in industries like manufacturing, retail, and supply chain
Their mission is to bring the power of foundational Vision AI models into the physical world, anywhere edge devices are deployed, through a full-stack platform that eliminates the need for in-house ML expertise
Visum AI offers a turnkey edge deployment solution (Vision 2.0) that handles all technical complexities—including model architecture selection, compression (pruning, quantization, distillation), latency/memory optimization, domain fine-tuning, plus system enhancements like Retrieval-Augmented Generation (RAG), Cache-Augmented Generation (CAG), and agentic tool use
Their platform is designed for use in industries where intelligent automation on edge devices can bring measurable value—especially in manufacturing, retail, and supply chain operations where quality, speed, and autonomous reasoning matter
No ML expertise is needed—Visum AI’s turnkey platform simplifies deployment for end customers, helping them focus on results rather than technical back-end challenges
To explore further or schedule a demo, Visum AI encourages contacting them via email (contact@visum.ai) or booking a demo directly on their website
Copyright © 2025 Visum AI Inc