Prompt Engineering courses

• Understand how large language models process prompts, generate responses, and use tokens to represent text.
• Learn how context windows influence what an AI system can consider when producing an answer.
• Distinguish between instructions, contextual information, examples, and user-provided data within a prompt.
• Recognize the principal limitations of LLM-based systems, including hallucinations, ambiguity, incomplete context, and sensitivity to prompt formulation.

• Formulate clear and specific instructions that communicate the intended task and expected outcome to an AI system.
• Provide relevant context and background information so that generated responses are aligned with the user's objective.
• Use constraints, roles, examples, and explicit requirements to control the scope, tone, depth, and behavior of a response.
• Define structured output formats such as tables, lists, templates, schemas, and step-by-step responses to obtain more consistent results.

• Apply few-shot prompting by providing representative examples that demonstrate the desired reasoning pattern, style, or output.
• Decompose complex tasks into smaller and more manageable subtasks to improve reliability and reduce ambiguity.
• Use structured reasoning approaches to guide models through multi-stage analysis while keeping the requested output clear and verifiable.
• Apply self-critique, review, and iterative refinement techniques to identify weaknesses in an initial response and progressively improve its quality.

• Design prompts for drafting, rewriting, and improving professional emails, reports, proposals, presentations, and other workplace documents.
• Use AI to prepare meeting agendas, summarize discussions, extract action items, and transform notes into structured follow-up material.
• Develop prompting workflows for research, summarization, comparison, analysis, brainstorming, and routine knowledge-work activities.
• Create reusable prompt templates that standardize recurring business tasks and improve consistency across everyday workflows.

See also: AI Productivity courses →

• Construct prompts that help AI systems interpret tables, spreadsheets, structured datasets, and descriptive business information.
• Extract relevant facts, categories, patterns, trends, anomalies, and relationships from structured or semi-structured data.
• Generate analytical insights while clearly separating observations supported by the supplied data from assumptions or interpretations.
• Request structured analytical outputs such as summaries, comparison tables, KPI explanations, recommendations, and decision-oriented reports.

See also: Data Analytics courses →

• Use prompting techniques to generate code from functional requirements while specifying programming language, constraints, dependencies, and expected behavior.
• Design prompts for debugging by providing error messages, execution context, relevant code, and reproducible information.
• Use AI to propose unit tests, test cases, edge cases, documentation, comments, and technical explanations for software artifacts.
• Apply structured prompts to code review and refactoring tasks in order to assess readability, maintainability, correctness, and potential improvements.

See also: AI for Developers courses →

• Design prompts for reviewing literature, identifying themes, extracting research information, and comparing findings across multiple sources.
• Transform documents and research material into structured representations such as evidence tables, summaries, taxonomies, and thematic syntheses.
• Use iterative prompting workflows to move from broad exploration to focused analysis, synthesis, and research-oriented outputs.
• Apply citation-aware prompting practices that require the model to distinguish source-supported claims from interpretation and to preserve traceability to provided material.

• Design prompts that combine textual instructions with images, documents, screenshots, diagrams, and other visual information.
• Formulate questions that guide multimodal AI systems to extract, compare, interpret, summarize, or reason about information contained in different media.
• Develop prompting strategies for document analysis, including requests involving tables, figures, layouts, and mixed textual-visual content.
• Understand how audio and other multimodal inputs can be incorporated into AI workflows and how instructions should be adapted to the capabilities and limitations of each modality.

• Write precise system and task instructions for AI agents that must plan actions, use tools, and complete multi-step workflows.
• Design prompts for Retrieval-Augmented Generation systems so that responses remain grounded in retrieved documents and external knowledge.
• Apply context-engineering principles to determine which instructions, retrieved information, history, and task state should be presented to the model.
• Structure agent memory, tool-use instructions, intermediate checks, and completion criteria to improve reliability in complex workflows.

See also: AI Agents courses →

• Evaluate prompts systematically using criteria such as accuracy, relevance, consistency, completeness, robustness, and adherence to required formats.
• Identify hallucinations and unsupported outputs, and design prompting and verification strategies that reduce their impact in enterprise workflows.
• Understand prompt-injection and instruction-conflict risks and apply appropriate separation, validation, and guardrail principles when working with untrusted content.
• Address privacy, confidentiality, reliability, and governance requirements when prompts contain organizational or sensitive business information.
• Develop reusable enterprise prompt standards, testing procedures, versioning practices, and approval processes that support safe and consistent AI adoption.

See also: AI Security courses →