AI LEARNING / 03 / INTERMEDIATE
Generative AI
& agents
Follow the progression from generating a response to working within a tool-enabled system.
THE IDEA TO KEEP IN VIEW
Keep the model, its information sources and its authority to act as separate questions.
SUGGESTED READING ORDER
Build one idea
at a time.
Begin with the Transformer architecture, then examine retrieval and context. Use the agent workflow guide to understand how tools, feedback and permissions change the role of a language model.
AFTER READING
Identify where a proposed AI workflow gets its information and which tools can change an external system.
01 / LANGUAGE MODELS
Understand the Transformer architecture
Explore tokens, attention and position, and why a fluent response still needs verification.
02 / RETRIEVAL & CONTEXT
Connect the model with relevant knowledge
Compare long context and retrieval-augmented generation in document and knowledge tasks.
03 / AGENT WORKFLOWS
Follow language models into agent workflows
Trace planning, retrieval, tool use and result checking within a workflow shaped by permissions and human approval.
Check your understanding
If a model can retrieve a document and call a tool, what still needs to be specified? Consider source quality, tool scope, review points and the evidence required before an action.
FURTHER EXPLANATIONS
Deepen a specific idea.
Continue through the wider collection after the essential readings.
From Transformers to Hybrid Intelligence: Mamba, SSMs, and Next-Generation AI Architectures
Explore Mamba, state space models, and hybrid AI architectures that combine sequence processing, Transformers, RAG, and human review for manufacturing workflows.
From Bigger to Cheaper: Efficient Transformers and the Economics of AI
Explore how FlashAttention and mixture of experts address Transformer efficiency, and why memory, latency, and deployment costs matter for enterprise AI.
From Scale to Systems: The Future of Transformers Is Not Just Bigger Models
Explore why Transformer-based AI needs more than model scale: efficient inference, RAG, tools, trusted data, and governance connected to real enterprise work.
The Evolution of Transformers: From Language Translation to AI Agents
Trace Transformer evolution from the 2017 attention paper through BERT, generative models, vision, RAG, and AI agents, with implications for enterprise workflows.
The Transformer Eight: How One Google Paper Created the Modern AI Industry
Meet the eight authors of Attention Is All You Need and explore how the Transformer architecture influenced generative AI, startups, and enterprise…
AI Agent Workflow: From Business Goal to Action
Follow six AI agent workflow steps: understand a goal, plan, retrieve context, use tools, recommend or act, and evaluate the result with…