Upgrade to Pro — share decks privately, control downloads, hide ads and more …

Build with AI on Google Cloud Session #1

Build with AI on Google Cloud Session #1

This is session #1 of the 5-session online study series with Google Cloud, where we take you onto the journey learning generative AI. You’ll explore the dynamic landscape of Generative AI, gaining both theoretical insights and practical know-how of Google Cloud GenAI tools such as Gemini, Vertex AI and Imagen 3.

Event page: https://gdg.community.dev/events/details/google-gdg-seattle-presents-build-with-ai-on-google-cloud-session-1-discovering-genai/

Margaret Maynard-Reid

January 22, 2025
Tweet

More Decks by Margaret Maynard-Reid

Other Decks in Technology

Transcript

  1. Build with AI on Google Cloud Session #1 Intro to

    GenAI 1/22/2025 Seattle | Surrey | Vancouver | Burnaby
  2. GDG Seattle 2 Margaret Maynard-Reid Yenchi Lin Clive Boulton Vishal

    Pallerla I/O Extended 2019 2024 Build with AI DevFest Seattle 2018 DevFest Seattle 2022 WTM Lightning Talks 2018 Cloud Study Jam 2018 DevFest Seattle 2017 DevFest Seattle 2016 DevFest Seattle 2015 DevFest Seattle 2024 Follow GDG Seattle on LinkedIn
  3. GDG Vancouver Follow GDG Seattle on LinkedIn 4 Follow GDG

    Vancouver on LinkedIn Join our GDG Vancouver Community Volunteer Interest Form
  4. Build with AI on Google Cloud Agenda • Study series

    overview • Cloud Skills Boost walkthrough • CSB beginner path repris • Intro to GenAI on Google Cloud • Prompt Design • Q & A - 20 min Seattle | Surrey | Vancouver | Burnaby
  5. Topics for this session • Online study series overview •

    Intro to GenAI on Google Cloud • GenAI beginner path • … 8
  6. Study series overview Follow 5 generative AI paths on Google

    Cloud Skills Boost: 1. 1/22/25 - Beginner: Intro to GenAI (link) 2. 2/5/25 - Generate Smarter GenAI Outputs (link) 3. 2/19/25 - Build & Modernize Apps with GenAI (link) 4. 3/5/25 - Integrate GenAI into Your DataFlow (link) 5. 3/19/25 - Deploy & Manage GenAI Models (link) Topics are not limited to the above. Each session: 2 short talks (by Googlers or experts) + Q&A section. 10
  7. What is a learning path? A learning path has multiple

    courses Each course has videos, recommended reading, quiz & hands-on labs. You will have at least two weeks to work through the materials It’s OK if you don’t finish and feel free to study ahead 11
  8. Access to Cloud Skills Boost • Sign up here: https://www.cloudskillsboost.google/

    • By RSVP, you get free access for a few months • The videos are accessible by default while labs each require a credit • You can work on each GenAI paths before or after each session Note: Make sure to sign up on Google Cloud Skills Boost with the same email that you used for event RSVP. 12
  9. 5 courses in the learning path 17 1. Intro to

    Generative AI (45 minutes) 2. Intro to LLMs (1 hour) 3. Intro to Responsible AI (30 minutes) 4. Prompt Design in Vertex AI (3 hours 45 minutes) 5. Responsible AI: Applying AI Principals with Google Cloud (2 hours)
  10. 1. Intro to generative AI • AI vs ML vs

    deep learning • What is Generative AI? LLMs and generative Image models… • Supervised vs unsupervised learning 18
  11. 5. Responsible AI: Applying AI Principles with GCloud 1. Be

    socially beneficial 2. Avoid creating or reinforcing unfair bias 3. Be built and tested for safety 4. Be accountable to people 5. Incorporate privacy design principles 6. Uphold high standards of scientific excellence 7. Be made available for uses that accord with these principles 22
  12. AI/ML GDE (Google Developer Expert) 3D artist Fashion Designer Instructor

    of MSIS, UW Foster Ex MS Design Studio, MSR, MS Bing About me margaretmz.art 24
  13. What is Generative AI? A type of AI that creates

    new content with generative models: 26 Text Image Video Audio Generative AI Text Image Video Audio
  14. What are Generative Models? “Generative models: take a machine, observe

    many samples from a distribution and generate more samples from that same distribution”. - Ian Goodfellow 2016 27
  15. What is an LLM? LLMs Explained [...] [...] [...] [...]

    0.02 0.03 0.9 0.01 0.0 … Dogs Rain Drops Fish Wind … and cats raining It’s
  16. Type of Generative Models • 2014 Generative Adversarial Networks (GANs)

    • 2016 Autoregressive Models • 2019 Variational autoencoders (VAEs) • Flow-based models • 2020 Diffusion models • 2022 Diffusion Transformer 29 Source: Lilian Weng blog (link)
  17. Diffusion Models 1. Gradually add gaussian noise to training data

    2. Learn how to reverse the process to generate images from noise. 30 Source: Nvidia developer blog (link) Forward image diffusion Generative reverse denoise
  18. CLIP: Contrastive Language-Image Pre-training CLIP is a bridge between NLP

    and computer vision, connecting text and Images It has a text encoder and image encoder, trained with 400 million image-text pairs. • DALLE, DALLE-2 • Stable Diffusion • Imagen, Imagen 2, Imagen 3 Paper: Learning Transferable Visual Models From Natural Language Supervision 31
  19. Diffusion Transformer Paper: Scalable Diffusion Models with Transformers SoTA models

    using diffusion transformer: • Pixart-a • SORA • Stable Diffusion 3 32
  20. The U-Net 33 U-Net architecture (image source: U-Net paper) 2015

    paper for medical imaging segmentation Used in many popular GenAI models: • Pix2Pix • CycleGAN • Diffusion Models (DDPM) • DALL-E • Midjourney • Stable Diffusion…
  21. U-Net vs Diffusion Transformer U-Net - not crucial to the

    good performance of the diffusion model - struggle with capturing long-range dependencies and global context in the input data Diffusion Transformer - More flexible - Can use more training data and larger model parameters - Transformers can model long-range dependencies without the need for deep networks or large filters, because of the self-attention mechanisms 34
  22. Thank you! Connect with me learn more about AI, art

    & design! @margaretmz @margaretmz @margaretmz @margaretmz 36
  23. Build with AI on Google Cloud Prompt Design - By

    Priti Y. 37 What is Prompt Design?
  24. Prompt Design Workflow 38 Define Objectives and Expected Outcomes Create

    the Prompt Content, Structure, Components Test the Prompt Identify Areas for Improvement Refine Iterate Iterate
  25. The Components of a Prompt • Define the Task •

    Contextual Information • Include Few-Shot Examples • System Instructions 39
  26. Common Pitfalls in Prompt Design • Vague Instructions • Not

    Providing Enough Context • Poorly Structured Prompts • Not Using Examples • Not Assigning a Role or Persona • Not Breaking Down Complex Tasks • Not Specifying the Output Format • Not Iterating on Prompts • Not Experimenting with Parameters 40
  27. Best Practices for Designing Prompts • Assign a Role •

    Structure Prompts • Instruct the Model to Explain its Reasoning • Break Down Complex Tasks • Experiment with Parameter Values • Specify the Output Format • Optimize Image Prompts • Tailor for Task Type • Consider the Model’s Limitations 41
  28. Thank you! Let's connect to explore AI Projects/Research and Board

    Games together! 43 @pritiyadav888 @mentors/priti-yadav
  29. Have fun studying! Action items: • Join discord - post

    your questions there • Get access to Cloud Skills Boost credits • Complete Beginner GenAI path on CSB • Get started on the 2nd GenAI path on CSB Next session: • Feb 5, 2025 - Session #2 - Deep Dive (RSVP) 46