Concepts

This page (WIP) summarizes the basic concepts learned while doing the hands-on QS demos.

TOC

• 2 NNs (NN, CNN, TF). The core matrix math classification mechanisms that are the basis of all AI.
• 2b Model. The packaged version of Internal Agent (procedural system) + Matrix math engine that work together to create AI “simulated intelligence” capabilities.
• 3 External agent (Custom or Framework). The (external) Procedural System + (Matrix math engine or packaged Model).
• 3b Project. Real world demo implementations of AI assistance for existing SW.

AI project components

2 NNs

The NN is the core of AI “intelligence”. A NN provides a pattern matching algorithm. The the NN is controlled by Python code (“agent”).

• 2.0 Modern architectures are evolutionary (not revolutionary)
• 2.1 NN (Inference)
• 2.3 Training
• 2.4 Convolution (CNN)
• 2.5 Transformer (TF)
• 2.6 Comparison of NN, CNN, and TF architectures
• 2.7 Why scaling works

2.0 Modern architectures are evolutionary (not revolutionary)

The latest architectures are not introducing anything radically different. I recently saw a video where one of the author’s of the original paper “Attention is all you need” (attention is a core functionality of LLM transformers) stated that no other ground-breaking architecture has appeared since the paper was pubished. So the concepts learned from basic demos apply to the latest foundation models.

2.1 Concepts – NN (inference)

NN demo D2ccc overview (first draft diagram generated by GPT)

2.3 Concepts – Training (NN)

Training input/output must match exactly what is desired in inference. The following code defines the training steps (more details later) for Tiny NN demo (demo D2ccc).

Training demo D2ccc overview (first draft diagram generated by GPT)

2.4 Concepts – Convolution

CNN demo D4 overview

2.5 Concepts – Transformer

2.6 Comparison of NN, CNN, and TF architectures

(make this a simplified version of 2.2b CNN<>TF comparison (WIP))

2.7 Why scaling works

2b Model concepts

The agent and the NN are typically packaged into a model that has an API that makes it possible for existing software to access the model via API. The model code that controls the TF NN is the “internal agent” (iAgent). Discusses

• 2b.1 What is a Model?
• 2b.2 Internal Agent (iAgent)
• 2b.3 APIs
• 2b.4 Local vs Cloud Models
• 2b.5 Building Models
• 2b.6 Modifying Models

This is a first draft (junk) diagram generated by GPT

3 Agent concepts

External agents access the model via API. This supports (1) reliable workflows built around models, tools, and automation and (2) tolerance of AI faults and unpredictable outputs. This section discusses

• 3.1 Agent Concepts
• 3.2 Code-first Agents
• 3.3 Agent Frameworks

3b AI project concepts

The real goal of AI projects is to integrate AI into an existing SW app. “Spinning up” real-world projects quickly with minimal code analysis or manual coding. Projects are where all the previous pieces come together. They aren’t just “more demos” — they’re the integration point.

• 2 teaches how models work.
• 2b teaches how models are packaged and served.
• 3 teaches how agents use models.
• 3b shows how agents solve real problems by integrating with existing software and external systems.

There are 2 kinds of Ai projects:

• PART 1: Enterprise SW demos. An existing application is the primary system. The AI is integrated to assist the application’s users and workflows.
• Part 2: Standalone AI Applications. The AI agent is the primary application. It may use external APIs, databases, files, or enterprise software as tools or data sources.

PART 1: Enterprise SW demos

Integrate AI into existing software (AI assistant).

The 4 stages (S1-S4) for ESW demos

S1 The traditional UI and ESW (#1). You use a browser to access the system.

S2 Native/Managed AI (#4). This is easiest to do with AI. An existing AI agent is connected to the target system. You configure the connection rather than build the agent.

For example

• S2a Claude + MCP → Slack ✅ (current demo)
• S2b Claude Tag in Slack ✅ (future demo)

S3 Integrated AI agent (#3).

• Your agent works INSIDE the target.
• Official plugin, extension, app, webhook, API integration, etc.
• Users interact through the target application’s UI.

S4 External AI agent / works OUTSIDE the target (#2).

• Users interact with your external agent rather than the target application’s UI.
• The target application is unaware of your agent.
• External AI. Most difficult to setup.

3b Project ecosystem for stages S1-S4 (add S4 DB?)

Part 2: Standalone AI Applications

Standalone AI applications have no existing enterprise application to integrate with. The external agent is the main application.

For example:

• 3b.3 NMAP Security Assistant. Accesses the local PC (target system). This is not only a practical tool, but also can help you learn very important security topics for your home PC/network setup. Nmap has real system, real data, security, MongoDB, OpenAI explanation, Mattermost alerts. It demonstrates nearly every important concept in one project.

3b standalone AI app (S4 only)

26.0628 (v1 26.0611)