This curriculum gives you an overview of what you need to learn to build AI-powered apps. It is meant for aspiring AI engineers, as opposed to AI researchers. So this guide is does not teach you to build and train neural networks, but rather to evaluate and productize pre-trained AI models.
- What is an AI model?
- What types of models exist?
- What is a Large Language Model (LLM)?
- How to evaluate different models
- Tools and playgrounds for evaluating models
- Inference
- Understanding stateless models
- System message
- Fine-tuninig
- Tokens
- Temperature
- Stop sequence
- Zero-shot
- Few-shot
- Frequency & presence penalty
- What is a vector?
- What is an embedding?
- What problem(s) do embeddings solve?
- How to create text embeddings
- Text splitting
- Vector databases (Chroma, Pinecone, Weaviate)
- Semantic search
- Long-term memory
- OpenAI functions
- Chain-of-thought prompting
- Reasoning & Acting (ReAct)
- Plan-and-executre