🧑‍💻 What is “AI Engineering?”

AI Engineering refers to the industry-relevant skills that data science and engineering teams need to successfully build, deploy, operate, and improve Large Language Model (LLM) applications in production environments.

In practice, this requires understanding aspects of both prototyping and production deployments. During the prototyping phase, Prompt Engineering, Retrieval Augmented Generation (RAG), and Fine-Tuning are all necessary tools to be able to understand and leverage. E.g.;

From The New Stack and Ops for AI by OpenAI on November 6, 2023.

When productionizing LLM application prototypes, there are many considerations to keep in mind when it comes to ensuring helpful, harmless, honest, reliable, and scalable solutions for your customers or stakeholders.

This course aims to cover the entire span of concepts and code necessary to prototype with LLMs as well as to deploy, operate, and improve them in production environments. We focus on tools that are both seen as industry-standard and are leading the way at the open-source edge in 2024.

This course combines aspects from our two previous courses, LLM Engineering and LLM Operations 👇

📜 AI Engineering Learning Targets

Prototype LLM Applications
- Building RAG Applications (e.g., Vector DBs, LLMs, Embedding Models)
- Fine-Tuning LLMs & Embedding Models (e.g., Hugging Face, LoRA, Quantization)
- Building Agent/Reasoning Applications (e.g., ReAct, function calling, external tools)
- LLM Application Deployment (e.g., Chainlit, Hugging Face)
Production-Grade Applications
- RAG Evaluation and Improvement (e.g., RAGAS, advanced retrieval techniques)
- Monitoring and Visibility Tooling (e.g., WandB, LangSmith)
- Efficient Inference and Serving (e.g., LangServe, prompt caching)
- Scaling LLM Applications (e.g., Amazon SageMaker)

🤔 Prerequisites

This course is designed for both aspiring AI Engineers and AI Engineering Leaders. The former will, of course, be more interested in coding everything themselves, while leadership positions typically require only an understanding of high-level concepts, tools, and infrastructure.

We have had many students succeed by taking both paths through the course.

The bare minimum prerequisites are:

Working knowledge of how to run Python code in Jupyter notebooks
A functional understanding of how machine learning and deep learning models are trained.

If you are still not at this level, you should with the Machine Learning Specialization from Deeplearning.ai. It will get you up to speed on both aspects.