NEXUSTEK

Agentic AI

Definition: Select and execute actions independently to achieve specific outcomes, without requiring constant human direction. Agentic AI can make decisions and execute tasks. Have agency, meaning they have goals and context, not prompts. Can handle multiple steps across enterprise systems.

Additional Context: Intiates action, executes a workflow, highly autonomous.

AI

Definition: AI is the application of advanced analysis and logic-based techniques, including machine learning, to interpret events, support and automate decisions, and take actions.

AI Enablement

Definition: Organizational or infrastructural process of preparing systems, teams, and environments to leverage AI effectively. This is a strategic concept.

Additional Context: Includes: governance, training, inegrating, building compliance, building platforms. Synonymous with AI Adoption, AI Integration, AI Operationalization.

AI Model

Definition: A computer program or algorithm that has been trained on a large dataset to recognize patterns, make predictions, or generate content without being explicitly programmed for every specific outcome.

Additional Context: GPT, Gemini, etc.

AIOps

Definition: The combination of big data and machine learning to automate IT operations processes using predictive analysis, including: Observability, Event Correlation, Monitoring, Anomaly Detection, Root Cause. AIOps is a strategy for managing modern IT environments to reduce manual effort, increase response.

Additional Context: Examples are automated detection and response actions, predictive analytics, alert correlation, root cause log correlation and noise reduction.

AI Operations

Definition: This is the steady state managment of AI Systems, models, agent and applications. Can be done by people or by systems.

Additional Context: This could be manual monitoring and remidaition of performance degradation, adjusting parameters, managing versions and rollbacks or allocating GPU resources.

AI Agent

Definition: A goal-driven software entity that uses AI techniques to complete tasks and achieve outcomes autonomously.

Additional Context: Behaves similarly to Agentic AI, but not neccesarily as complex as Agentic AI. Agentic AI is often used to refer to a platform, or capabilities.

AI Application

Definition: Software systems that embed AI capabilities to deliver intelligent advice, automate decisions, generate content, or execute tasks traditionally performed by humans.

Additional Context: Common examples are virtual assistants like Siri, or Alexa. A business example might be using AI in hiring to scan resumes and rank candidates.

Deploying AI

Definition: Covers the actions that allow AI Agents or Applications to be activated or turned into Production. This is a tactical set of activities.

Additional Context: Activating the model, providing data access, connecting to platforms, etc. Synonymous with Deploying AI, Implementing AI, Provisioning AI, Configuring AI.

Gen AI, Generative AI

Definition: AI techniques that learn a representation of artifacts from data and use it to generate brand-new, unique artifacts that resemble but don’t repeat the original data.

Additional Context: Generative AI, Gen AI and GenAI are synonymous in usage. Responds to prompts, generates content, non-autonomous.

Knowledge Base

Definition: A knowledge base is a centralized, organized collection of facts, rules, relationships, and heuristics that an AI system uses to simulate understanding, reasoning, or decision-making. It can include structured data (like ontologies or taxonomies), unstructured data (like documents or articles), and semantic relationships between concepts.

Additional Context: Content types could incude FAQs, install guides, glossaries, policies or workflows.

Large Language Model (LLM)

Definition: An advanced type of model trained on a vast amount of data to generate human type responses. It uses deep learning to perform tasks.

Additional Context: GPT, Gemini.

Managed Intelligence

Definition: Managed Intelligence Services are focused on lifecycle management of business workflows and particularly AI Agents, Applications and Models shifting from reactive management to proactive management. Managed Intelligence Services extends the offer from traditional Professional and Managed services via the continuous insight, automation, and refinement across these business workflows using proactive intelligence. We monitor performance, maintain reliability, and refine outputs to ensure decisions stay fast, accurate, and aligned with goals.

Additional Context: Outputs include performance refinement, hallucination reduction, compliance adherence, patching, upgrade compatibility.

RAG

Definition: Retrival Augmented Generation. Instead of relying solely on pre-trained knowledge, a RAG system first retrieves relevant documents or data from an external knowledge source (like a database or search index), and then uses a generative model (such as a transformer-based LLM) to produce a response based on both the retrieved content and the input query.

Additional Context: A user might ask a question "How do I…" The RAG then queiries available data sources and presents back in human language "This is how you do…"

Workflow

Definition: Workflow, in either a general IT or AI specific context, refers to a defined sequence of tasks, operations, or processes that are executed to complete a specific business or technical objective. It often involves automation, integration across systems, and rules-based execution to ensure consistency, efficiency, and scalability.

Additional Context: The step by step process or logic used to complete the objective.