Ava Meredith

Projects

Visit my Github repository

Visit my blog, Head in the Clouds, for detailed How-To guides

Featured Projects

AWS Amplify AI Chatbot

AI Chatbot can see past conversations and resume conversations. Secure login with Amazon Cognito. This project showcases the development of an AI chatbot application leveraging AWS Amplify and ReactJS. The chatbot provides secure authentication and persistent chat history storage.
GitHub Repository: Amplify AI Chatbot Project

Agents on AWS with Langchain

This project demonstrates the deployment of a Llama 3.1 8b Instruct model as an AI chatbot assistant. The architecture consists of four main layers: model deployment, LangChain integration, agent tools, and user interface. The foundation is built on AWS SageMaker, where the Llama model runs on a g5.2xlarge instance. LangChain serves as the integration framework, handling the communication between the model and various tools. The agent layer incorporates Wikipedia for knowledge retrieval and a calculator for math operations, while managing conversation memory and JSON responses. The user interface is built with Streamlit, providing a clean, web-based chat experience. The system uses conversation buffer memory to maintain context and formats responses in JSON for consistent handling. This architecture enables natural conversations while leveraging external tools for enhanced capabilities.
Jupyter Notebook: LangChain Agent

Multimodal AI Examples

This project demonstrates bakery website content generation, utilizing Anthropic's Claude V2 for text generation of menu items and Stability AI's Stable Diffusion XL for image creation. It generates weekly baked goods menus with corresponding images, implementing specific model parameters for output control. The second example leverages the Claude 3 Sonnet model for advanced image analysis, demonstrating various prompting techniques, including chain of thought, and processing base64 encoded images.
These examples highlight the integration of multiple AI models within Amazon Bedrock for diverse applications in content generation and image analysis.
Jupyter Notebook: Multimodal

Develop a Django App on AWS and deploy it with Elastic Beanstalk

This project showcases a full stack, dynamic ice cream shop web application built using Django, a high-level Python web framework. Developed entirely in the AWS Cloud9 integrated development environment, the application leverages cloud-based tools. The ice cream shop app features a user-friendly interface allowing customers to browse flavors and create orders. On the backend, it utilizes Django's ORM for database management. For deployment, the application utilizes AWS Elastic Beanstalk. Elastic Beanstalk simplifies the deployment process, automatically handling capacity provisioning, load balancing, and application health monitoring.
GitHub Repository: DjangoApp

Resources

Visit my blog, Head in the Clouds. I blog about technology, especially about AWS Cloud, AI and ML.

About

Hi, I'm instructor Ava! 👋 Let me tell you a bit about my journey. Before joining the awesome team at AWS, I spent a decade as a college professor, helping students level up their tech skills. I taught Python, Java, Data Science and Cloud Computing. I've also got tech industry experience under my belt, having worked at a couple PNW Tech companies. I'm a graduate of St. Martin's University, where I earned my BS in Computer Science and MBA. In November 2022, I joined AWS as a Technical Trainer. These days, I'm all about AI. It's cutting-edge stuff, and I love every minute of it! I have eight AWS certs, including Authorized Instructor Champion, AI Practitioner, ML Engineer Associate Early Adopter, AWS SysOps. When not teaching or building I enjoy hiking and spending time with my family and my 2 adopted cats.

Contact

Elements

i = 0;

while (!deck.isInOrder()) {
    print 'Iteration ' + i;
    deck.shuffle();
    i++;
}

print 'It took ' + i + ' iterations to sort the deck.';