Generative AI & LLM Engineering — Skills Plan

Start with these published lessons

These existing AI & ML lessons cover foundations for this path. They count in their original track.

1. 01 · FoundationsWhat is Generative AI?Compare generative and discriminative models and learn where major generative model families fit.
2. 02 · FoundationsTransformers and Self-AttentionUnderstand the transformer building blocks behind modern language models.
3. 03 · FoundationsTokenisation and Word EmbeddingsFollow how text is tokenised and represented for language-model processing.
4. 04 · FoundationsLLMs — Pretraining, RLHF, and Scaling LawsStudy next-token pretraining, instruction alignment, and scaling concepts.
5. 05 · FoundationsContext Windows and Model LimitsBudget instructions, history, retrieved evidence, and output; diagnose context overflow and missed information.
6. 06 · FoundationsInference Parameters and Model SelectionCompare model and generation settings using task quality, format reliability, latency, and cost.
7. 07 · Core skillsPrompt EngineeringPractice prompt patterns and compare how task instructions affect model outputs.
8. 08 · Core skillsRAG — Retrieval-Augmented GenerationConnect retrieval and generation to answer questions using external documents.
9. 09 · Core skillsLLM Agents and Tool UseSee how a model can select tools and use their results as part of a workflow.
10. 10 · AdvancedLLM Fine-Tuning in PracticeExplore fine-tuning workflows and when adapting model weights is appropriate.
11. 11 · AdvancedAdvanced RAGInvestigate retrieval improvements and the trade-offs in more complex RAG systems.
12. 12 · SpecialisationsMultimodal Models — CLIP and LLaVAExplore models that process and relate information across text and images.

• ◆Python for AI applications

  Build the Python for AI applications knowledge needed to follow the Generative AI & LLM Engineering foundations.

• ◆HTTP APIs and JSON

  Build the HTTP APIs and JSON knowledge needed to follow the Generative AI & LLM Engineering foundations.

• ◆Git and virtual environments

  Build the Git and virtual environments knowledge needed to follow the Generative AI & LLM Engineering foundations.

• ◆AI and machine-learning foundations

  Build the AI and machine-learning foundations knowledge needed to follow the Generative AI & LLM Engineering foundations.

• ◆Linear algebra and probability essentials

  Build the Linear algebra and probability essentials knowledge needed to follow the Generative AI & LLM Engineering foundations.

• ◆Natural language processing fundamentals

  Explain how Natural language processing fundamentals works and where it fits in the Generative AI & LLM Engineering learning path.

• ◆Transformer architecture and attention

  Explain how Transformer architecture and attention works and where it fits in the Generative AI & LLM Engineering learning path.

• ◆Generative models and LLMs

  Explain how Generative models and LLMs works and where it fits in the Generative AI & LLM Engineering learning path.

• ◆Tokens and tokenization

  Explain how Tokens and tokenization works and where it fits in the Generative AI & LLM Engineering learning path.

• ◆Context windows and model limits

  Explain how Context windows and model limits works and where it fits in the Generative AI & LLM Engineering learning path.

• ◆Inference parameters and model selection

  Explain how Inference parameters and model selection works and where it fits in the Generative AI & LLM Engineering learning path.

• ◆Prompt design and versioning

  Use Prompt design and versioning to implement a working capability and verify its behavior in Generative AI & LLM Engineering.

• ◆Structured outputs and validation

  Use Structured outputs and validation to implement a working capability and verify its behavior in Generative AI & LLM Engineering.

• ◆Provider APIs and streaming

  Use Provider APIs and streaming to implement a working capability and verify its behavior in Generative AI & LLM Engineering.

• ◆Embeddings and semantic search

  Use Embeddings and semantic search to implement a working capability and verify its behavior in Generative AI & LLM Engineering.

• ◆Vector databases

  Use Vector databases to implement a working capability and verify its behavior in Generative AI & LLM Engineering.

• ◆Retrieval-augmented generation (RAG)

  Use Retrieval-augmented generation (RAG) to implement a working capability and verify its behavior in Generative AI & LLM Engineering.

• ◆LLM agents and tool calling

  Use LLM agents and tool calling to implement a working capability and verify its behavior in Generative AI & LLM Engineering.

• ◆Reranking and hybrid retrieval

  Evaluate Reranking and hybrid retrieval trade-offs against the performance, reliability, and safety needs of Generative AI & LLM Engineering systems.

• ◆Fine-tuning concepts and RAG trade-offs

  Evaluate Fine-tuning concepts and RAG trade-offs trade-offs against the performance, reliability, and safety needs of Generative AI & LLM Engineering systems.

• ◆LLM evaluation and regression testing

  Evaluate LLM evaluation and regression testing trade-offs against the performance, reliability, and safety needs of Generative AI & LLM Engineering systems.

• ◆Responsible AI and governance

  Evaluate Responsible AI and governance trade-offs against the performance, reliability, and safety needs of Generative AI & LLM Engineering systems.

• ◆Guardrails and output validation

  Evaluate Guardrails and output validation trade-offs against the performance, reliability, and safety needs of Generative AI & LLM Engineering systems.

• ◆Security, privacy, and prompt injection

  Evaluate Security, privacy, and prompt injection trade-offs against the performance, reliability, and safety needs of Generative AI & LLM Engineering systems.

• ◆Latency, reliability, and cost management

  Evaluate Latency, reliability, and cost management trade-offs against the performance, reliability, and safety needs of Generative AI & LLM Engineering systems.

• ◆Deployment, monitoring, and observability

  Evaluate Deployment, monitoring, and observability trade-offs against the performance, reliability, and safety needs of Generative AI & LLM Engineering systems.

• ◆Grounded knowledge assistant

  Deliver a working outcome with Grounded knowledge assistant, then test it against the project requirements.

• ◆Document extraction pipeline

  Deliver a working outcome with Document extraction pipeline, then test it against the project requirements.

• ◆Evaluated support-drafting assistant

  Deliver a working outcome with Evaluated support-drafting assistant, then test it against the project requirements.

• ◆LLM system design

  Explain your approach to LLM system design, defend key trade-offs, and discuss what you learned.

• ◆RAG trade-offs

  Explain your approach to RAG trade-offs, defend key trade-offs, and discuss what you learned.

• ◆Evaluation and safety reviews

  Explain your approach to Evaluation and safety reviews, defend key trade-offs, and discuss what you learned.

• ◆Portfolio walkthrough

  Explain your approach to Portfolio walkthrough, defend key trade-offs, and discuss what you learned.