Start Here

This learning hub approaches AI from three connected perspectives: building AI systems, designing their architecture, and leading their responsible adoption.

You do not need to study every topic in the same order. Choose the path closest to your current goal, then use the other pillars when your work requires a broader perspective.

← Back to Home

Choose Your Primary Path

If you want to…	Begin with…	You will focus on…
Build and operate AI applications	Build AI Systems	Backends, LLM applications, RAG, evaluations, agents, and production reliability
Make defensible technical and architectural decisions	Design AI Architecture	Requirements, constraints, data flow, interfaces, system boundaries, reliability, and trade-offs
Guide responsible organizational AI adoption	Lead AI Transformation	Strategy, management, change, governance, value, operating models, and transformation

If you are unsure where to begin

Start with the path that matches the decisions you need to make today. You can add a supporting path without abandoning your primary one.

How the Three Pillars Connect

AI Strategy & Transformation defines direction

It asks why AI should be used, which outcomes matter, whether the organization is ready, how value will be measured, and what governance is required.

AI Engineering Design & System Architecture shapes the system

It translates goals and constraints into components, interfaces, data flows, reliability requirements, and explicit architectural trade-offs.

AI Engineering builds and operates the system

It turns the architecture into working software, tests whether the system behaves as intended, and produces evidence that informs future design and strategy decisions.

These are not isolated disciplines. Strategy without engineering can become unrealistic, engineering without architecture becomes fragile, and architecture without organizational context can solve the wrong problem.

Path 1 — Build AI Systems

Choose this path if you want to create reliable AI-powered applications and understand the software engineering beneath them.

What you will learn

Technical foundations — practical Python, files, configuration, errors, testing, and developer tooling
Backend foundations — HTTP, REST, asynchronous programming, APIs, validation, and FastAPI
LLM application foundations — prompt design, structured outputs, model selection, and failure modes
Data and retrieval — SQL, PostgreSQL, embeddings, vector databases, indexing, and retrieval
RAG systems — ingestion, chunking, retrieval, generation, grounding, and citations
Evaluations — test datasets, retrieval metrics, quality criteria, and regression testing
Agents and tools — tool calling, workflows, human approval, failure recovery, and MCP
Production concerns — logging, observability, latency, cost, security, deployment, and reliability

Useful foundations

Comfortable Python fundamentals
Basic command-line and Git usage
A working understanding of data structures and APIs
Essential statistics and machine-learning concepts when a system requires them

Machine learning is treated here as supporting knowledge rather than the destination. The focus remains on designing, building, evaluating, and operating AI applications.

Begin with: AI Engineering Roadmap
Explore the pillar: AI Engineering

Path 2 — Design AI Architecture

Choose this path if you want to reason clearly about systems before committing to tools, frameworks, or implementation choices.

What you will learn

Engineering-thinking foundations — problems, symptoms, assumptions, constraints, and trade-offs
Requirements — user needs, functional requirements, non-functional requirements, and success criteria
System decomposition — responsibilities, components, boundaries, coupling, and cohesion
Interfaces and data flow — contracts, state, behavior, control flow, and movement of data
Architecture patterns — selecting patterns because they fit the problem rather than because they are fashionable
Reliability and change — failure modes, observability, scalability, maintainability, and technical debt
AI application architecture — model services, retrieval, orchestration, evaluation, guardrails, and human oversight
Architecture decisions — documenting alternatives, consequences, reversibility, and evidence

Useful foundations

Experience reading or building small software systems
Willingness to define the problem before choosing technology
Basic familiarity with how APIs, databases, and application components interact

Begin with: Engineering Design & System Architecture Roadmap
Explore the pillar: AI Systems Design and Architecture

Path 3 — Lead AI Transformation

Choose this path if you want to help an organization select, govern, adopt, and scale AI responsibly.

What you will learn

AI opportunity framing — identifying valuable problems instead of beginning with tools
Organizational readiness — data, skills, processes, leadership, culture, and technology foundations
AI strategy — priorities, portfolios, roadmaps, investment choices, and alignment with organizational goals
AI management — capabilities, teams, operating models, vendors, delivery, and value measurement
AI change management — stakeholders, communication, participation, training, resistance, and workflow redesign
AI governance — accountability, policies, risk classification, oversight, responsible AI, and human review
AI transformation — pilots, adoption, process redesign, maturity, scaling, and continuous improvement
Technical literacy for leaders — understanding enough engineering and architecture to challenge assumptions and make informed decisions

Useful foundations

Experience with teams, projects, programs, operations, or organizational change
Interest in connecting technical possibilities to real organizational outcomes
Willingness to examine incentives, risk, adoption, and governance alongside technology

Begin with: AI Strategy & Management Roadmap
Explore the pillar: AI Strategy and Transformation

Combine the Paths Deliberately

Your primary path determines where you spend most of your time. A supporting path gives you the context needed to make better decisions.

AI Engineer: Build AI Systems + Design AI Architecture
AI Architect: Design AI Architecture + Build AI Systems
Engineering leader: Design AI Architecture + Lead AI Transformation
AI transformation leader: Lead AI Transformation + technical and architectural literacy
AI consultant: Lead AI Transformation + the engineering depth required by the engagement

Avoid trying to complete all three paths simultaneously. Use one as the main curriculum and draw from another when a project exposes a real need.

How to Learn from This Garden

Use the notes actively rather than reading them as isolated reference material:

Understand the mental model and vocabulary.
Predict what a system, process, or piece of code should do.
Trace how information, decisions, and failures move through it.
Modify an example, assumption, requirement, or design choice.
Build something small enough to expose gaps in understanding.
Reflect on trade-offs, evidence, organizational impact, and what should change next.

The goal is not to collect tools. It is to develop the judgment required to build useful AI systems, design them responsibly, and help organizations adopt them successfully.

Recommended Entry Points

← Return to Home

Educational material that may include the author's interpretations or opinions and may be developed with AI assistance. Read the public disclaimer.