fr
en
Developing generative AI applications on AWS
Your AI use cases need appropriate services and evaluated outcomes. With AWS, connect data, models and integration to assess the technical options more clearly. Develop a framework for designing experiments and assessing implementation requirements.
- Duration
- 2 days 14 hours
- Code
- AWS07FR Code
Presentation
To drive innovation in generative AI, AWS offers Amazon Bedrock, which simplifies access to foundation models, and Amazon CodeWhisperer, a valuable coding assistant. These services enable businesses to create intelligent applications and increase productivity. AWS also offers an extensive range of AI services to enhance generative AI capabilities, covering areas such as image analysis and speech transcription.
This intensive 2-day course equips you with the essential practical skills to master generative AI and its supporting AWS services. You will explore real-world use cases, examine prompt engineering principles in depth and address critical responsible AI considerations. Through a series of interactive hands-on labs, you will learn to design effective prompts, evaluate results meaningfully and strategically plan the implementation of future generative AI projects.
By the end of this comprehensive programme, you will have developed the expertise to understand, apply and harness generative AI responsibly and effectively within the AWS ecosystem. You will master key concepts, best practices and essential tools such as Amazon Bedrock and LangChain to make full use of this transformative technology within your organisation.
Objectives
By the end of this generative AI on AWS course, you will be able to:
- Master fundamental generative AI concepts, including its importance, benefits, risks and key terminology;
- plan a generative AI project by identifying the business value of use cases, explaining the stages and managing risks;
- understand how Amazon Bedrock works, its benefits, typical use cases, architecture and cost structure, and perform a practical demonstration;
- apply prompt engineering principles and techniques, including basic and advanced methods, specific model selection and bias mitigation;
- develop generative AI applications by identifying their components and customising foundation models (FMs) through their APIs and inference parameters;
- integrate LangChain with large language models (LLMs) to build complex applications, using chains, chat models, embeddings, document loaders, retrievers and agents;
- implement generative AI architectures and apply these concepts to build and test use cases with Amazon Bedrock, LangChain and retrieval-augmented generation (RAG);
- identify AWS services for monitoring, securing and governing Amazon Bedrock applications.
Program
Module 1: Understanding generative AI (the art of the possible)
- What is machine learning (ML)?
- The basics of generative AI and its fundamental concepts.
- Generative AI use cases.
- The risks and benefits of AI.
- Generative AI in practice: an overview.
Module 2: Planning and contextualising an AI project
- The business context for generative AI.
- Key steps in planning an AI integration project.
- Identifying and mitigating risks specific to generative AI projects.
Module 3: Getting started with Amazon Bedrock
- Introduction to the Amazon Bedrock service.
- Architecture and use cases.
- Best practices for using Amazon Bedrock.
- Configuring access and using playgrounds (demo).
Module 4: Mastering prompt engineering fundamentals
- Foundation model basics.
- Prompt engineering fundamentals.
- Basic and advanced prompting techniques.
- Refining a basic text prompt (demo).
- Model-specific prompting techniques.
- Managing prompt misuse.
- Bias mitigation.
- Mitigating image bias (demo).
Module 5: Building Amazon Bedrock application components
- Applications and use cases.
- Core application components.
- Foundation models and the FM interface.
- Using datasets and embeddings.
- Additional application components.
- Retrieval-augmented generation (RAG).
- Model fine-tuning.
- Securing generative AI applications.
- Generative AI application architecture.
- Word embeddings (demo).
Module 6: Using Amazon Bedrock foundation models
- Introduction to foundation models.
- Using Amazon Bedrock FMs for inference.
- Amazon Bedrock methods.
- Data protection and auditability.
Lab:
- Invoke an Amazon Bedrock model for text generation using a zero-shot prompt.
Module 7: Integrating the LangChain framework
- Optimising LLM performance.
- Integrating AWS and LangChain.
- Using models with LangChain.
- Building prompts.
- Structuring documents with indexes.
- Storing and retrieving data with memory.
- Using chains to sequence components.
- Managing external resources with LangChain agents.
Module 8: Applying architecture patterns
- The basics of architecture patterns.
- Text summarisation.
- Question answering.
- Chatbots.
- Code generation.
- LangChain and agents for Amazon Bedrock.
Labs:
- Use Amazon Titan Text Premier to summarise text in small files.
- Summarise long texts with Amazon Titan.
- Use Amazon Bedrock for question answering.
- Build a chatbot.
- Use Amazon Bedrock models for code generation.
- Build conversational applications with the Converse API.
Audience
This course is intended for:
- Software developers interested in using large language models without needing to fine-tune them.
Prerequisites
The following prerequisites are recommended for this AWS course:
- Completion of the AWS Technical Essentials course;
- Intermediate proficiency in Python.
Teaching and assessment methods
- Initial skills assessment
- Training materials provided to participants
- Continuous assessment throughout the course
- End-of-course feedback questionnaire
- Combination of theory and practical application
- Attendance records
- Post-course follow-up evaluation
- Quiz / multiple-choice questions
- Practical exercises
Course highlights
- AWS-certified expert trainers: learn from recognised, AWS-certified trainers with in-depth knowledge of generative AI practices and AWS tools and services for developing and deploying applications based on large language models.
- Interactive hands-on learning: master generative AI tools and techniques on AWS through concrete demonstrations and practical labs. Prepare to address real-world challenges in prompt engineering, LangChain integration, Amazon Bedrock usage and the implementation of generative AI application architectures in the cloud.
- Key skills development: carefully designed course content helps you acquire essential skills in designing and implementing generative AI solutions, managing large language models and LangChain, and efficiently managing AWS resources.
Dates and sessions
Choose the date and delivery format that suit you.
No upcoming sessions are currently available.
AWS, Amazon Bedrock, Amazon Titan and other AWS marks are registered trademarks of Amazon.com, Inc. or its affiliates.