1. Grounding
Grounding is also known as retrieval-augmented generation (RAG). Grounding makes sure that the AI-generated content is specific to your organization.
Example:
Prompt will search the Knowledge articles setup and send it to LLM along with the Prompt. LLM generates responses with the Prompt and grounding info(relevant articles).

2. Token to Characters calculation
In OpenAI, one token generally corresponds to ~4 characters of text for common English text.
For Example,
Test – 1 token
Testable – 2 tokens
https://platform.openai.com/tokenizer

3. Why do we use Prompt Templates?
The Prompt Templates are used to generate prompts. The generated Prompts are sent to Large Language Models (LLMs) in order to generate content.

Prompt Templates allow us to have placeholders to fill the data dynamically.

4. Supervised Learning
Generative AI is built using supervised learning.

Supervised learning is the technology that has made computers very smart when given an input to generate a corresponding output.

Supervised learning is a category of machine learning that uses labelled datasets to train algorithms to predict outcomes and recognize patterns. Unlike unsupervised learning, supervised learning algorithms are given labelled training to learn the relationship between the input and the outputs.

5. What is retrieval-augmented generation (RAG)?
Retrieval-Augmented Generation (RAG) is a technique that grants generative artificial intelligence models information retrieval capabilities. It modifies interactions with a large language model (LLM) so that the model responds to user queries with reference to a specified set of documents, using this information to augment information drawn from its own vast, static training data. This allows LLMs to use domain-specific and/or updated information.Use cases include providing chatbot access to internal company data or giving factual information only from an authoritative source.
LLMs responses will be so good or sometimes it won’t be. RAG is an AI framework which improves the LLMs responses by grounding with external sources of knowledge. It helps to optimize the LLM response.

6. AI Waves
Wave 1: Predictive
Wave 2: Generative
Wave 3: Agent

7. Agent AI Orchestration
AI agent orchestration is the process of managing and coordinating multiple AI agents to work together to achieve a task. Agent AI Orchestration involves ensuring that the agents communicate, share data, and take decisions collaboratively.

8. What is AI Orchestration?
AI orchestration is the process of coordinating different AI tools and systems so that they can work together effectively.

9. AI Agents
AI agents are the individual AI systems or entities that are designed to perform specific tasks independently.

10. LangChain
LangChain is a software framework that helps facilitate the integration of large language models (LLMs) into the applications. As a language model integration framework, LangChain’s use-cases largely overlap with those of language models in general, including document analysis and summarization, chatbots, and code analysis.

11. What is AI MCP?
The Model Context Protocol (MCP) is an open-source standard developed by Anthropic to seamlessly connect AI systems with various data sources, such as content repositories, business tools, and development environments.

12. gPRC(Remote Procedure Calls)
gRPC is a cross-platform high-performance remote procedure call framework. gRPC was initially created by Google, but is open source and is used in many organizations. gRPC, unlike REST, was designed specifically to allow developers to create high-performance APIs for microservice architectures across distributed data centers. It’s better suited for internal systems that require real-time streaming and large data loads.

13. What is an LLM Gateway?
An LLM Gateway is a centralized platform that helps in the deployment and management of LLMs(Large Language Models). It gives us a unified interface to connect the applications with one or more LLMs. We will have good control over what is passed to and received from the LLMs. It’s a good Auditing mechanism.

14. Multi-Modal LLM
Multi-Modal LLM has the ability to process multiple inputs like text, files, audio, etc unlike traditional LLM which can process only text.

15. AI Agents Vs. Agentic AI
AI Agents
AI agents will perform tasks but they still require human inputs. They follow the configured or programmed instructions.
Example:
Autopilot mode in cars, Apple Siri, Google Assistant, etc.
Agentic AI:
Agentic AI is different from AI Agent. It’s a system of AI agents that operate without human interventions.