Prompt Engineering for Multimodal AI Training in Bangkok, Thailand

This training equips participants to design prompts that interact with multi-modal AI systems capable of processing and generating text, images, audio, and code. The course focuses on mastering prompt formats for tasks such as document analysis, visual generation, speech-to-text transcription, and multi-modal chaining.

Core Prompting Across Modalities

  Overview of multi-modal AI systems (ChatGPT, DALL·E, Sora)

  Differences in prompting for text, image, and audio inputs

  Prompt syntax and formatting best practices for each modality

  Text-to-image prompting: specificity, style, and composition

  Image-to-text: describing visuals and extracting information

  Speech-to-text with Whisper: prompt-based refinement of transcripts

  File and image upload prompting techniques (e.g., document Q&A, OCR)

Advanced Chaining and Real-World Applications

  Multi-step/multi-modal prompt chains (text → image → text)

  Prompting across tools for tasks like creative design, accessibility, analysis

  Role prompting in multi-modal scenarios (e.g., “Act as a design reviewer”)

  Real-world workflows: product content creation, UX/UI critique, meeting summaries

  Use-case design: marketing, education, legal, design, and healthcare

  Ethical considerations, limitations, and prompt transparency in multi-modal systems

Hands-On Activities