As someone deeply interested in the world of machine learning and AI, I’ve found myself drawn to the teachings of Andrew Ng. Over the past few months, I’ve completed seven of his courses, each offering unique insights into the evolving landscape of machine learning and AI. Here’s my take on these courses and how they’ve shaped my understanding:
- Machine Learning Specialization
- ChatGPT Prompt Engineering for Developers
- Generative AI for Everyone
- Building system with chatGPT API
- Neural Networks and Deep Learning
- Functions, Tools, and Agents with LangChain
- Vector Databases: from Embeddings to Applications
1. Machine Learning Specialization – A Strong Foundation
This beginner-friendly 3-course specialization provides a solid foundation in machine learning concepts and applications. The courses move from supervised learning algorithms like regression and classification to more advanced techniques like unsupervised learning and reinforcement learning. I appreciated the hands-on coding exercises in Python and found it accessible even with limited programming experience. This is one of the best introductory machine learning courses out there.
2. Generative AI for Everyone – Demystifying the Hype
Targeted at a non-technical audience, this course demystifies generative AI, explaining its applications and limitations. The content is well-organized, balancing the pros and cons and dispelling common myths. It’s a great course for anyone looking to understand generative AI from conception to launch, including building effective prompts.
3. Neural Networks and Deep Learning – Understanding Key Models
This course started to go a bit deeper into deep learning, how it is applied to supervised learning, major categories of models (CNNs, RNNs, etc.), and when they should be applied. Taking this course after the “Machine learning specialization” can feel a bit redundant with certain content. Feel free to skip them.
4. When you feel ready to build your first Gen AI application
When you feel that you are ready to start building your first application, especially a Gen AI application, start with the following courses “ChatGPT Prompt Engineering for Developers”, “Building system with chatGPT API”,
These courses will save you a lot of time as they will give you an overview of the most important steps you should think about when building your applications. Of course, they come with sample code too so they are very practical.
ChatGPT Prompt Engineering for Developers
This course is a boon for both beginners and advanced learners. It is taught by Andrew and also Isa Fulford from OpenAI. It focuses on the nuances of prompt engineering and the use of Large Language Models (LLMs). The course brilliantly elucidates the core concepts of LLMs and offers practical insights into best practices for prompt engineering. It’s particularly valuable for those looking to understand and leverage the power of LLM APIs in various applications. For example, how to ask GPT to output its answer in JSON format so that you can use it later on in your application.
Building Systems with the ChatGPT API
A follow-up to the Prompt Engineering course, this one-hour session is perfect for beginners in LLMs. It provides hands-on examples and teaches how to efficiently build multi-step systems using large language models. The course is an excellent primer for those interested in prompt engineering and large language model applications. I appreciate the guidance on the below with actual code examples:
- How to use “Moderation API” to validate user input, avoid prompt injection
- Chain of thought reasoning
- Chaining prompts
- Output evaluations
5. Ready to dive even deeper?
After these courses, if you want to go deeper into Gen AI application building, the two courses “Functions, Tools, and Agents with LangChain” and “Vector Databases: from Embeddings to Applications” are perfect.
Understanding what a vector database is, how it works, and how it can help you build a multi-modal generative AI application is super cool.
As I continue my journey into machine learning, I’m grateful for the breadth of topics covered by Andrew’s courses. They’ve given me a well-rounded perspective and actionable skills to build, apply, and responsibly scale AI systems. I look forward to leveraging these learnings in real-world projects ahead.
Chandler