AWS Builder Center
What I Learned from AWS Skill Builder: Fundamentals of Machine Learning and Artificial Intelligence

What I Learned from AWS Skill Builder: Fundamentals of Machine Learning and Artificial Intelligence

A practical reflection on the fundamentals of AI, ML, deep learning, and generative AI, along with the AWS services that support building, deploying, and scaling intelligent applications.

Full-Stack Developer
Recently, I completed the Fundamentals of Machine Learning and Artificial Intelligence course on  AWS Skill Builder .
The course helped me build a stronger foundation in AI, Machine Learning, Deep Learning, and Generative AI, while also understanding how AWS services support these technologies in real-world applications.

AI, ML, Deep Learning, and Generative AI

One of my key takeaways was understanding how these technologies relate to each other:
  • Artificial Intelligence (AI) is the broader field of creating systems capable of human-like intelligence.
  • Machine Learning (ML) enables systems to learn patterns from data rather than being explicitly programmed for every scenario.
  • Deep Learning (DL) uses multi-layered neural networks to learn complex patterns.
  • Generative AI uses deep learning models to generate content such as text, images, audio, video, and code.

Machine Learning Fundamentals

I learned that data is at the foundation of machine learning. A typical ML workflow can be represented as:
 Data → Data Preparation → ML Algorithm → Training → Model → Evaluation → Inference 
The course covered labeled and unlabeled data, as well as structured data such as tabular and time-series data and unstructured data such as images, video, and text.
I also learned about the three major approaches to ML:
  1. Supervised Learning: Learning from labeled data.
  1. Unsupervised Learning: Finding patterns in unlabeled data.
  1. Reinforcement Learning: Learning through rewards and penalties.
An important lesson was that ML development is iterative. Models need to be tested, evaluated, and improved continuously.
I also learned about inference, including batch inference for processing data in groups and real-time inference for generating predictions as inputs arrive.

Deep Learning

The course introduced the fundamentals of neural networks, including nodes, input layers, hidden layers, and output layers.
Deep learning is particularly useful for complex data and applications such as computer vision and natural language processing.

Generative AI and Foundation Models

Generative AI introduced me to Foundation Models (FMs), which are pre-trained on large and diverse datasets and can be adapted for multiple tasks.
I learned about the Foundation Model lifecycle:
 Data Selection → Pre-training → Adaptation → Evaluation → Deployment → Feedback 
The adaptation stage introduced techniques such as:
  • Prompt Engineering: Using instructions, context, input data, and output indicators to guide a model.
  • Retrieval-Augmented Generation (RAG): Providing external information to a model without changing its underlying weights.
  • Fine-tuning: Further training a model for specific tasks or requirements.
I also explored LLMs, tokens, embeddings, vectors, diffusion models, GANs, and VAEs.

AWS Services for AI and ML

Another valuable part of the course was understanding which AWS services support different AI and ML requirements.
  • Amazon SageMaker for building, training, and deploying ML models.
  • Amazon Bedrock for accessing a choice of high-performing foundation models and building Generative AI applications.
  • Amazon SageMaker JumpStart for accessing models and ML solutions.
  • Amazon Q and Amazon Q Developer for generative AI assistance.
  • Amazon Translate, Textract, Polly, Lex, Comprehend, Kendra, and Personalize for specific AI capabilities.
  • AWS DeepRacer for exploring reinforcement learning.
I also learned that AI/ML solutions require consideration of performance, scalability, deployment, and cost, not just the technology itself.

Key Takeaway

Overall, this course gave me a better understanding of the end-to-end AI/ML lifecycle, from data preparation and model development to evaluation, deployment, inference, and continuous improvement.
More importantly, I gained a clearer understanding of how AWS provides the infrastructure, managed services, and tools needed to turn AI and ML concepts into practical applications.
Any opinions in this article are those of the individual author and may not reflect the opinions of AWS.
Enjoyed reading this content? Let the author know!

Your likes, comments, shares, and saves help creators reach more builders.

Loading recommendations

Loading article