Fuelling Innovation with Gen AI on AWS: Discovery Workshop

Our Generative AI Discovery workshop provides you with a comprehensive understanding of Generative AI and its significance for the modern enterprise. We will explore the immense capabilities of Large Language Models (LLMs) and provide insights on how to pick the right model for different use-cases under different constraints. We’ll go through several techniques for exploiting LLMs, including prompt engineering, chain-of-thought reasoning and Retrieval Augmented Generation (RAG) architectures.

We’ll also go through guardrails, defensive development techniques and architectures that you can employ to use generative AI in a safe and secure manner. Additionally, we’ll cover available services and tools for Generative AI, with a particular focus on AWS’s offerings in this space. We’ll explore various use-cases such as Chatbots, Enterprise Search, Recommendation Systems, Content Generation and Developer tools. Our collaborative whiteboarding sessions ensure interactive learning and idea sharing. Most importantly, we empower you with clear next steps and an action plan, ensuring you leave the workshop ready to apply your knowledge effectively.

Prior to workshop

Before commencing the Discovery workshop, we request a 1-hour meeting to tailor the workshop to your unique use-cases and specific needs. We can also tailor the agenda based on the audience, i.e. we can span the range from a business-focused to a technical hands-on workshop.

Day of the workshop

We would recommend the workshop to be in-person, but can adjust to a remote audience if needed. On the day of the workshop, we will deep dive into relevant aspects of Generative AI and your identified use cases. We will employ physical and digital whiteboarding techniques to go through different topics. Depending on the audience, we might also conduct hands-on development sessions. We will also introduce the array of AWS services specifically designed for AI/ML, including vector databases, languages translation and comprehension, storage, and compute services. These critical components serve as the foundation for creating a robust, end-to-end, scalable Gen AI powered application on AWS. Finally, we will brainstorm on how your organization can move forward with one or more use-cases on generative AI.

Post-Session Deliverables

Following the immersive workshop, we will provide your organisation with a clear strategy for exploiting the power of Gen AI, in alignment with your overarching business objectives. This includes a high-level architectural overview of the prioritised use cases, an estimation of the time and effort investments required, and a strategically phased approach towards achieving success.

Duration: 8 hours

Sample Agenda (Can be customized to Customer requirements):

Introduction to Gen AI
A brief history of Gen AI and its current state
Large Language Models
Introduction to different models and their trade-offs
How to use LLMs (prompts, chains, agents, etc)
Gen AI Use-Cases
Chatbots, Search, Question-Answering, Content Generation, Code Generation, etc
Gen AI Architectures
Retrieval Augmented Generation (RAG) Patterns
Chatbot-related Patterns
Content & Code Generation-related Patterns
Common components - Vector DBs, Orchestrators, MLOps, LLMOps
AWS Gen AI Services Overview
AI Services on AWS (Amazon SageMaker, Amazon Bedrock etc.)
Vector Databases (Amazon OpenSearch, Amazon Kendra etc.)
Gen AI Demos on AWS
Hands-On Session (Optional )
Governance & Safeguards
Guardrails
Security & Privacy Considerations
Customer Use Cases & Current State Deep Dive
Breakout Sessions for Each Use Case
Collaborative White-boarding
Next Steps and Action Planning