Quali integrates NVIDIA AI Enterprise to simplify Agentic AI deployment

New integrations lower the barrier to adoption for Agentic Artificial Intelligence (AI) by automating management across development resources, including Large Language Model (LLM) (LLMs), data services, Graphics Processing Unit (GPU) infrastructure, and more to deliver high-performing AI agents.

Quali, a provider of platform engineering solutions for infrastructure automation and management, announced its integration with NVIDIA AI Enterprise software, including NVIDIA NIM™ microservices and NVIDIA AI Blueprints, to simplify the creation and management of Agentic AI solutions.

Agentic AI represents an opportunity for enterprises, impacting both internal operations and customer experience.

By leveraging NVIDIA NIM microservices and NVIDIA AI Blueprints included in the NVIDIA AI Enterprise software platform, the Quali Torque Software-as-a-Service (SaaS) platform simplifies the orchestration and management of each layer of the Agentic AI tech stack: accelerated infrastructure, cloud services and data pipelines, LLMs and AI models, and AI agents and applications.

With Torque providing unified orchestration, lifecycle management, and cost optimization, each layer of the development stack supports one another to ensure reliability, accuracy, and efficiency.

To accomplish this, Torque manages the entire infrastructure lifecycle using Environments as Code (EaC)—a model that transforms cloud resources into fully managed, self-service environments supporting mission-critical operations, such as AI workloads, software development, demos, training, and more.

Quali can deploy and manage the entire tech stack supporting Agentic AI solutions as a stateful, dynamic environment through the integration of NVIDIA technology. This approach automates routine tasks, enabling organizations to adopt Agentic AI faster, scale efficiently, and concentrate on innovation rather than infrastructure management.

Key highlights of this release include:

• Easy-to-Use Modules of NVIDIA AI Enterprise & Other NVIDIA Resources: Torque creates reusable modules defining each component needed to deliver an AI agent, including NVIDIA accelerated compute, NVIDIA NIM microservices, and pre-trained models and data science frameworks included with NVIDIA AI Enterprise. This normalization enables no-code orchestration of each stack layer supporting the AI solution, thereby accelerating delivery and reducing the barriers to adopting Agentic AI.
• AI-Driven Environment Design, Creation, & Reusability: Torque’s AI Copilot leverages reusable modules to design and generate new environment blueprints in response to user-submitted prompts. Additionally, Torque’s graphical environment design tool allows users to drag-and-drop resources and set dependencies visually. As users adjust the environment design, Torque modifies the code in the environment blueprint file automatically, further reducing the need for complex coding. Once complete, Torque saves this blueprint as a reusable file that can be deployed, maintained, and monitored continuously.
• Simplified Provisioning & Maintenance of Individual Layers of the Agentic AI Tech Stack: Torque executes code to provision each layer of the AI tech stack, continuously monitors the state of those resources, and notifies users about anomalies, including infrastructure errors, configuration drift, and other unexpected updates. This allows Torque administrators to tailor the user experience, enabling proactive reconciliation of issues.
• Streamlined Integration of Each Layer of the AI Tech Stack: Once provisioned as a managed environment in Torque, each AI tech stack layer can be published for other users to access as input. For example, a developer can select the live GPU clusters, cloud-based data services, and AI models needed from a simple pick list in Torque’s provisioning experience. Torque leverages those inputs to provision the AI agent while also enabling users to maintain the live environment supporting each layer of the stack.
• Automating Critical Tasks for AI Performance: Torque workflows can define and automatically execute routine tasks needed to maintain high-performing AI solutions, such as training and data quality assurance, eliminating manual work while providing visibility for users.
• Dynamic GPU Scaling in Response to Application Needs: As AI models supporting the agent transition through phases, Torque automatically scales GPUs up and down to provide adequate computing capacity for resource-intensive tasks while scaling down for less intensive tasks.

Lior Koriat, Quali CEO, stated, “Complexity has always been at the core of the problems we solve for our customers and partners. As more organizations look to embrace AI, the ability to cut through complexity is the key to delivering the kinds of AI experiences that customers expect. We’re thrilled to develop a streamlined approach for delivering AI solutions leveraging NVIDIA AI, and we look forward to helping our community unlock these opportunities.”