Projects
Assignments
Overview
The Generative AI Engineer (Associate) course offers a foundational understanding of generative AI concepts and Alibaba Cloud technologies. Participants will explore the principles of generative AI, delve into machine learning and deep learning techniques, and gain hands-on experience with Alibaba Cloud’s AI services. This course is designed to equip learners with the skills to identify and describe appropriate generative AI services and solutions, playing a key role in their organization’s AI initiatives.
Objectives
By the end of this course, leaner will be able to:
-
Understand the fundamentals of generative AI, including its development history and key concepts.
-
Gain insights into machine learning, deep learning, and natural language processing techniques.
-
Explore Alibaba Cloud’s generative AI offerings, such as Qwen and Model Studio.
-
Learn about model training, optimization, and deployment strategies.
-
Develop prompt engineering skills to interact effectively with generative models.
Prerequisites
This course is designed for beginners; no prior experience with generative AI or Alibaba Cloud is required.
Course Outline
- Introduction to Generative AI Concepts
- What is GenAI
- Development history of GenAI
- GenAI vs traditional AI
- Key concepts in GenAI
- Use cases and industry applications
- Logistics, education, healthcare, etc.
- Core foundational models
- GANs, VAEs, and Transformers
- What is GenAI
- Foundational Technologies Behind Today’s AI Capabilities
- What is machine learning
- Understanding machine learning
- Training techniques (supervised, unsupervised, and
reinforcement learning) - Types of ML algorithms and their applications
- What is deep learning
- Biomimicry in deep learning
- Understanding RNNs and CNNs
- What is NLP
- How machines can understand language
- The NLP development pipeline
- NLP at work
- Tokenization, embeddings, and language models
- What is machine learning
- Introduction to Qwen
- What is Qwen
- Model capabilities and performance
- Qwen model offerings
- Deploying Models on Alibaba Cloud
- Introduction to GPU-accelerated ECS instances
- Introduction to PAI
- Introduction to ComputeNest
- Model Studio Fundamentals
- What is Model Studio
- Concepts & components
- System prompts
- Multi-model comparison
- Agent and RAG creation
- Understanding Language Models
- Evolution of language models
- Scaling laws vs performance
- Pre-training Generative Models
- Data collection & preparation
- Model architectures
- Tuning techniques (instruction and alignment tuning)
- Generative Model Optimization
- Understanding augmentation techniques
- Understanding retrieval augmented generation systems (RAGs)
- Understanding agents
- Multi-agent workflows
- Prompt Engineering Fundamentals
- Understanding how generative models “think”
- What is prompt engineering
- The “genius in a room” analogy
- Crafting prompts:
- The five elements of a good prompt
- “Training” LLMs through effective prompting
- Basic prompting techniques
- Instructions
- Examples (few-shot)
- Role-play
- Chain-of-thought
