Published: Jun 30, 2025

NCS Sunshine.Operations

To date, AI Agents powered by Large Language Models (LLMs) and Agentic AI have significantly advanced in their ability to comprehend and contextualise enterprise knowledge. These agents are increasingly becoming the “engineers” and “companions” of IT operations teams that are capable of planning, reasoning and recommending actions across AI and machine learning (ML) workflows. Given that AI/ML operations are both time-sensitive and resource-intensive, the adoption of such intelligent companions introduces a layer of autonomy that reduces operational overhead and minimises the need for human intervention in repetitive tasks. This shift not only drives down costs but also empowers enterprises to redirect focus toward strategic initiatives and innovation.

Sunshine.Operations is an AIOps platform designed to automate incident triage, system log analysis and operational task flows. The platform is embedded with NCS-curated knowledge and designed to work within enterprise constraints, boosting productivity while aligning with security, data and workflow governance.

In collaboration with NVIDIA, NCS has adopted NVIDIA AI Enterprise integrated with NVIDIA solutions such as NVIDIA NIMs, NVIDIA Retriever and NVIDIA AI Blueprints software for enterprises to enhance the performance of the following Sunshine.Operations solutions:
    A. NCS Agentic AI Deployment Agents 
    B. NCS Agentic AI Management Agents
    C. NCS Guardian Network Agents
 

NCS has developed an Agentic AIOps IaC (Infrastructure as Code) Deployment solution which merges advanced autonomous intelligence with strategic human oversight to revolutionise cloud infrastructure deployment. This solution harnesses advanced Agentic AI components to power every phase of IaC deployment, as follows:

• LLM-Based IaC Code Generation: Using state-of-the-art LLMs, the system translates user requirements into robust IaC code, ensuring configurations align with exact infrastructure needs.
• Multi-Modal Agentic Code Generation: The tool converts AWS architecture diagrams into actionable text prompts. An intelligent IaC developer agent then refines these prompts to generate precise, compliant IaC scripts.
• Autonomous Execution with Terraform: The agent-driven workflow integrates smoothly with Terraform, automatically running terraform init, plan and apply. This end-to-end automation reduces the need for manual input while still enabling human approval at critical decision points.
• Dynamic Standardisation and Compliance: The agent automatically retrieves naming conventions and project-specific details, suggesting corrections for non-compliant resource identifiers and ensuring consistent provisioning.
• Agentic Troubleshooting and Optimisation: The solution’s autonomous diagnostic engine identifies deployment errors in real-time, performs web-based research for analogous issues, and recommends remediation steps — all enhanced by human validation when needed. 

This solution equips organisations with a scalable, consistent and secure framework for autonomously deploying and managing cloud infrastructures, leveraging Agentic AI to dynamically optimise environments while preserving critical human expertise.
 

The Agentic AIOps Infrastructure Management solution is designed to transform how engineers diagnose and resolve issues within virtual machines (VMs) and Kubernetes clusters. This innovative solution fuses advanced LLM-driven agents with Retrieval-Augmented Generation (RAG) technology, enabling a highly autonomous and context-aware troubleshooting process — augmented of course, with human oversight for decision-making.

Key technical highlights:

• LLM-Driven Assistance: Intelligent agents interpret user-defined prompts to deliver targeted troubleshooting support for VMs and Kubernetes clusters.
• Enhanced Diagnostic Context: By harnessing RAG, the platform retrieves comprehensive environment data and historical resolution records to enrich diagnostic insights
• Autonomous Data Collection: An agentic workflow employs secure remote execution protocols and orchestrated environment queries to systematically gather operational logs, system events and service metrics.
• Integrated Root Cause Analysis: The solution synthesises multi-source data to propose effective remediation strategies, automatically generating detailed root cause analysis reports that are fed back into the RAG system for continuous refinement. It has the capability to initiate corrective actions, closing the remediation loop without delay.

By merging autonomous diagnostics, deep contextual insights and proactive auto-remediation with strategic human oversight, we empower organisations to rapidly and accurately resolve complex issues to minimise downtime and maximise reliability.

Leveraging on NVIDIA AI Solutions for NCS Agentic AI Deployment Agents and NCS Agentic AI Management Agents:

These Agentic AIOps solutions use NVIDIA Retriever NIMs that are deployed via NVIDIA Blueprints and pre-packaged with seven NIMs (NeMo Retriever Llama 3.2 Embedding NIM, NeMo Retriever Llama 3.2 Reranking NIM, Llama 3.3 Nemotron Super 49B v1 NIM, NeMo Retriever Page Elements NIM, NeMo Retriever Table Structure NIM, NeMo Retriever Graphic Elements NIM and PaddleOCR NIM). The NVIDIA Retriever will serve as the context retriever at the heart of its Agentic AIOps solution, powering the RAG pipeline that ingests, vectorises, reranks and extracts structured knowledge from diagrams, tables and logs.
 

The NCS Guardian Network Agents help IT and security teams manage network threats by automatically detecting unusual activity, investigating the root cause and recommending specific actions to fix the problem. It handles common threats like DDoS attacks and network scans, so your team can respond faster and with greater accuracy. NCS Guardian Network Agents reduces the time that your teams spends reviewing false alarms and manually investigating incidents. It makes expert-level insights available to everyone, so even junior engineers can handle complex situations with confidence. By using NCS Guardian Network Agents, teams respond to threats faster, with fewer mistakes and in a consistent way regardless of who is on shift.

What it does when a threat is detected

When NCS Guardian Network Agents spots suspicious activity on the network, it:

• Identifies the issue clearly. For example, it can flag that the system is under a DDoS attack, or that someone is probing for open ports.
• Explains why. It shows the exact traffic patterns, IP addresses and data logs that triggered the alert, so teams can verify what’s happening.
• Recommends what to do next. It provides step-by-step instructions and scripts tailored to your setup, helping teams contain and resolve the issue quickly.

Leveraging NVIDIA AI Solutions for NCS Guardian Network Agents

NCS Guardian Network Agents is integrated with NVIDIA Inference Microservices (NIM) to power its Agentic RCA engine. The stack includes:

• Llama 4 Scout, a purpose-built LLM hosted via NVIDIA’s OpenAI-compatible API for reasoning, planning and remediation guidance.
• Built to support future multi-agent workflows and RAG pipelines as roadmap items.
   

NCS Sunshine.Operations demonstrates NCS’s advanced capabilities in AI for IT operations. By developing intelligent agents that automate deployment, infrastructure management and network threat response, NCS is pushing the boundaries of what AI can achieve.