AI SOLUTIONS ARCHITECTURE

Let’s connect! Meet your instructor, Toby Fotherby, Senior AI/ML Specialist Solutions Architect at AWS. Get familiar with the course and ask any questions you have. 

• Instructor Introduction
• Course Objectives & Flow
• Q & A

Explore the diverse responsibilities of an AI Solutions architect - and start to understand how these vary across companies, sectors and specialisations. Grasp the importance of collaborating with peers including data engineers and data scientists. 

• Role & Responsibilities of an AI Solutions Architect
• Peer roles
• Core skills & tools
• Exploring your goals 

Assignment #1: Self assessment and identifying focus areas for growth

Dive into fundamental AI/ML algorithms and concepts, covering key topics such as classification and regression techniques. Solidify your understanding of foundational concepts and get hands-on with a lab session to set up your AI/ML environment.

• Classification
• Regression
• Lab: Setting up your AI/ML environment

Assignment #2: Establishing your AI/ML development environment

Get to the essence of supervised and unsupervised learning, delving into core concepts and tools essential for model training. Get practical experience in training a regression model.

• Supervised and unsupervised learning
• Model Training: Core concepts & tools
• Lab: Training a regression model 

Assignment #3: Training a regression model: Predicting house prices

Lean in and focus on advanced concepts such as clustering, forecasting, recommendation engines and neural networks. Deepen your understanding of complex algorithms and their applications. Learn to train a clustering model, further enhancing your practical skills.

• Clustering 
• Lab: Training a clustering model 
• Forecasting & recommendation engines
• Neural networks, CNNs & Transformers

Explore Natural Language Processing and document processing techniques then engage in a lab session designed to consolidate your understanding. Delve into core concepts and tools for image processing to broaden your skill set in both textual and visual data analysis.

• Natural Language Processing
• Document processing
• Lab: Document processing 
• Image processing: Core concepts & tools

Assignment #4: Document processing

Uncover the significance of embeddings and their utility in various applications. Take a dive into vector databases, gaining insights into efficient data storage and retrieval techniques in high-dimensional spaces.

• What are embeddings and why are they useful?
• Sentence embeddings
• Lab: Comparing sentences using embeddings
• Vector databases

Explore and assess the landscape of quick wins and challenges within generative AI applications. Cover application architecture fundamentals and engage in a hands-on lab session focused on prompt engineering.

• Quick wins and challenges
• Application architecture fundamentals
• Lab: Prompt engineering

Assignment #5: Prompt engineering: sentiment analysis and entity detection

Get familiar with the fundamentals of application integration, essential for deploying generative AI solutions effectively. Participate in a lab session focusing on retrieval augmented generation techniques, and explore the development of GenAI chatbots and agents.

• Application integration: fundamentals
• Lab: Retrieval augmented generation
• GenAI chatbots
• GenAI Agents

Explore and engage. Explore the integration of cloud platforms and AI services to streamline model training and deployment processes. Leverage AI/ML tools available on cloud platforms for various tasks, and engage in a lab session focused on deploying models and conducting inference in a cloud-based environment.

• Cloud Platforms & AI Services
• AI/ML tools for model training and deployment
• Generative AI AI/ML tools for model training and deployment
• Lab: Cloud based model deployment and inference

Focus on how to achieve scalability and operational excellence in machine learning operations. Learn strategies and best practices to streamline ML workflows and ensure efficient and reliable operations throughout the model lifecycle.

• ML Ops, scalability and operational excellence
• Data pre-processing
• Model training
• Model evaluation
• Model deployment & monitoring

Navigate the principles and good practices crucial for successful AI/ML solutions. Explore what success entails in solution design and how to map requirements effectively to AI/ML components. Validate your designs and develop proof of concepts to ensure alignment with project objectives and stakeholders' needs.

• What does success look like?
• Mapping the requirements to AI/ML components
• Scale, pattern of use & level of detail
• Validating the design
• Developing a proof of concept

Assignment #6: Capstone project: Solution Proposal

Equip yourself with strategies for building resilient AI systems and optimising their performance and cost-effectiveness. Explore techniques for seamless model training and deployment, managing model and data drift, and enhancing inference processes. 

• Model training and deployment pipeline
• Model and Data Drift
• Inference serving & pipelines
• Cost Optimisation 

Assignment #7: Evaluating model price/performance

Address crucial aspects of data security and system scalability in AI applications. Understand how to secure data throughout the model training and deployment pipelines, as well as in application pipelines. Grasp the importance of factoring in scalability considerations, including handling big data in AI projects and facilitating interactive transactions at scale. 

• Data security
• Scalability
• Big data and AI
• Interactive transactions at scale & autoscaling

Assignment #8: Capstone project: Develop proof of concept

Critical topics surrounding AI include ethical considerations, regulations and bias in models. Explore the ethical implications of AI technologies and regulations governing their use. Investigate methods for detecting and mitigating bias to ensure fairness in AI model development and deployment.

• Ethical Considerations & Regulations
• Bias and Fairness in AI Models

Assignment #9: Capstone project: Solution Proposal - Develop proof of concept

Hone your skills in effectively communicating AI solutions to stakeholders. Get pitch perfect with presentation techniques tailored for showcasing AI solutions. Explore strategies for successful solution delivery, ensuring alignment with stakeholders' expectations and project objectives.

• Communicating AI Solutions to Stakeholders
• Presentation Skills for AI Solutions
• Demo: Presenting an AI Solutions Architecture
• Delivering solutions

Assignment #10: Capstone project: Solution Proposal - Wrapping it up!

Get ahead with current trends and future outlooks in AI/ML and Generative AI fields. Receive the guidance you need to tailor your CV and interview preparation for AI roles. Engage in Q&A sessions to showcase and discuss your capstone project, preparing you for success in your AI career journey.

• AI/ML and Generative AI: Trends and outlook
• Interview & Resume Guidance
• Ongoing Self-Education
• Capstone Project Q&A