Vianai Systems, the human-centered AI platform, and products company launched the Vian H+AI™ Platform for reliable, optimized enterprise-wide deployment, management, and governance of machine learning (ML) models at scale. The Vian H+AI Platform brings a human-centric approach, combined with advanced AI techniques to address and remove barriers that currently prevent widespread AI adoption into production environments. A lack of trust, transparency, and governance of AI models over their entire lifecycles, exacerbated by skyrocketing infrastructure costs, has significantly slowed the time-to-value of ML models – preventing enterprises from realizing the full potential of their AI investments. Vian is on a mission to change that.
The first set of capabilities available within the Vian H+AI Platform includes Vian MLOps for managing, optimizing, deploying, and governing ML models for production at scale. The Vian H+AI MLOps Platform provides a collaborative user experience across the AI workflow between data scientists, MLOps engineers, and risk auditors to operationalize and manage models as a continuous lifecycle over a long period. Vian Performance Optimization within the Platform dramatically accelerates model speed and throughput even on commodity hardware. In addition, the open, modular architecture of the Vian H+AI MLOps Platform enables MLOps teams to seamlessly leverage the entire end-to-end capabilities of the Platform or simply use components of it, to create a truly customized solution to fit their specific needs.
“AI plays a strategic role at Schneider Electric, in our products and services, and business operations. We have tested Vianai’s platform on our AI applications running in remote oil pumps, and observed very impressive AI model execution speed and throughput improvements on the same hardware,” said Dr. Peter Weckesser, Chief Digital Officer and Member of the Executive Committee, Schneider Electric. “Based on this outstanding performance, we are evaluating additional potential use-cases in Energy Management and Industrial Automation leveraging Vianai’s technology to enable us to deliver high-performance AI safely and scalably across our ecosystem.”
“Over the last three years, we have worked with dozens of the world’s largest and most respected companies, to learn the needs of businesses that AI must serve. Our H+AI platform marries those business needs with our advanced technical work, using the power of design thinking, and the result is several breakthroughs that create a new category of AI capabilities. Capabilities that early enterprise AI tools simply do not have,” said Dr. Vishal Sikka, Founder & CEO of Vianai Systems. “Our platform delivers AI that is high-performance, low-footprint and cloud-agonistic, and at the same time, transparent, reliable, and manageable across the lifespan of data, business activities, and models, at a global scale. We are now bringing this new class of AI capabilities, via our powerful platform and the extraordinary business benefits that our early customers have seen, to all enterprises.”
With the Vian H+AI MLOps Platform, enterprises can now build a foundation for human-centered AI systems. The Platform combines several open-source tools, Vian-proprietary techniques and optimizations, and a human-centric approach to bring AI to the enterprise at scale across diverse landscapes. With the Vian H+AI MLOps Platform, enterprises have maximum flexibility in model-building tools used by data scientists, can dramatically accelerate model performance without cost-performance tradeoffs, and have a central place to manage all models including end-to-end lifecycle management and monitoring for risks such as drift, uncertainty, bias, and explainability, seamlessly retraining models when and if needed. The Platform delivers these unique capabilities via its:
Unified User Experience: MLOps engineers can now quickly operationalize models, regardless of the tools used by data scientists to create those models. There is no longer a need to restrict data scientists to using specific proprietary tools, which currently slows the process of moving AI models into production.
Open, Modular Architecture: Vian MLOps is open source-based, built from scratch to deliver a flexible, componentized architecture. MLOps engineers can quickly and easily plug-and-play components and make decisions on the best cloud, database, repositories, and other components to use, without changing the API. In addition, the modular architecture enables MLOps teams to leverage the entire Platform end-to-end or seamlessly bring specific components of the Platform into their AI workflows.
Performance Optimization: With Vian Performance Optimization, a key capability within the Vian H+AI Platform, customers can dramatically accelerate the execution speed and throughput of models without costly hardware upgrades.
Risk Monitoring: Vian MLOps provides a rich set of model risk monitoring capabilities for data quality and integrity, drift, uncertainty, bias, explainability, and more. Leveraging integrated open-source as well as Vian proprietary technologies, each model can follow a customized risk monitoring plan before and after deployment to production.
Vian MLOps delivers the foundational starting point for building human-centered AI systems and reflects the company’s unique approach to creating solutions that address real business needs, based on empathy and understanding of the customer. In its strategic engagements with customers across a variety of industries, Vian saw a clear need for a more human-centered approach to AI, in particular, the need for a platform to manage and operationalize reliable, explainable, observable, and trustworthy ML models. Going forward, the Vian H+AI Platform will deliver a wide range of advanced AI tools and techniques, to help enterprises pursue their most strategic opportunities by leveraging human-centered AI.