Become Databricks Certified Generative AI Engineer Associate

Explore how to design and implement LLM enabled solutions on Databricks and prepare for the certification exam, including format, pricing, and prep resources and mock tests.

Navigate this course by audience type: seasoned generative AI learners can jump to quizzes and mock tests, while upskillers should follow the videos in sequence to prepare for the certification.

Explore how artificial intelligence learns from data, distinguish AI, machine learning, deep learning, and neural networks, and see how generative AI creates text, images, and videos using transformers.

Databricks launches a free edition for learning and AI exploration; create a free account, verify via email, and use notebooks, ETL pipelines, and MLflow to build LLM and rack-based applications.

Explore prompt engineering and prompts as inputs to LLMs, including multimodal instructions, one shot and few shot prompting, and a five-part framework: task, context, references, evaluation, and iteration.

Explore how context and specificity in prompts steer outputs, using FIFA World Cup examples and JSON-formatted results with winner and runners-up.

Explore few-shot prompting and chain-of-thought for more accurate ai outputs. Learn instruction-based prompting, react, tree-of-thought, and meta prompting, plus agentic refinement with practical examples.

Develop iterative prompt development to align generative AI outputs with goals by refining prompts, breaking long paragraphs, rephrasing, and adding constraints, including json format constraints.

Understand how tokenization turns text into tokens and IDs for llms, and compare word, character, and subword tokenizers. See how token types affect cost and performance.

Embeddings convert text into numerical vectors that place meaning in a vector space. They enable semantic search and rag systems, through input, tokenization, neural encoding, and cosine similarity.

Explore five areas of design applications, from crafting explicit prompts and structured outputs to selecting model tasks, building chain components, mapping inputs and outputs, and ordering tools for multi-stage reasoning.

Learn data preparation for RAG workflows: chunk by logical sections, remove extraneous content, OCR-extract scanned PDFs, and store text in Delta Lake via data frames in Unity Catalog.

Master application development for the Databricks generative AI engineer associate exam, covering PySpark with S3 and Rag workflow, Lang chain components, prompts, and safety and chunking considerations.