This comprehensive training course prepares you for the AWS Certified AI Practitioner exam – a new and highly popular Foundational-level certification from AWS.
The AWS Certified AI Practitioner course prepares learners to confidently understand and discuss artificial intelligence (AI), machine learning (ML), and generative AI—specifically within AWS contexts. This foundational-level course is built for professionals who may use AI/ML concepts but don’t necessarily build the models themselves. You’ll explore core AI/ML ideas, AWS services like SageMaker and Bedrock, and how to apply these technologies responsibly and effectively in real business contexts. The curriculum is structured to guide you through exam domains, such as foundational AI concepts, generative AI, ethical AI, and security—equipping you for the AIF-C01 certification exam.
Module 1: Fundamentals of AI & ML (Week 1–2)
(Aligned with Domain 1 – 20%)
1.1 AI Basics & Core Terminologies
- What is AI?
- What is ML?
- Deep Learning, Neural Networks, NLP, Computer Vision
- LLMs, Models, Algorithms
- Training vs Inference
- Bias, Fairness, Hallucinations
1.2 Types of AI/ML & Learning Approaches
- Supervised / Unsupervised Learning
- Reinforcement Learning
- Types of Data (Structured, Unstructured, Text, Image, Time-Series)
1.3 Practical Use Cases of AI
- Fraud detection
- NLP (Summaries, sentiment)
- Computer Vision
- Speech recognition
- Recommendation Systems
1.4 ML Lifecycle Overview
- Data collection
- Exploratory data analysis (EDA)
- Pre-processing
- Feature engineering
- Model training, tuning, evaluation
- Deployment & monitoring
1.5 AWS AI/ML Managed Services
- Amazon Comprehend
- Amazon Translate
- Amazon Transcribe
- Amazon Rekognition
- Amazon Textract
- Amazon Contact Center AI tools
Module 2: Generative AI Foundations (Week 3–4)
(Aligned with Domain 2 – 24%)
2.1 Generative AI Concepts
- Tokens, chunking, embeddings, vectors
- Foundation models
- Transformers
- Multi-modal models
- Diffusion models
2.2 Use Cases of Generative AI
- Text generation
- Chatbots
- Code generation
- Summarization
- Image/Video generation
- Translation
- Customer support bots
2.3 Capabilities & Limitations
- Advantages of GenAI
- Limitations (hallucinations, non-determinism, accuracy issues)
- Choosing the right generative model
- Business value of AI tools
2.4 AWS Generative AI Tools
- Amazon Bedrock (Core foundation)
- PartyRock – Bedrock Playground
- Amazon Q (Enterprise AI Assistant)
- Bedrock Model options:
- Amazon Titan
- Anthropic Claude
- Meta Llama
- Mistral
- Cohere
Module 3: Applications of Foundation Models (Week 5–6)
(Aligned with Domain 3 – 28%)
3.1 Designing Foundation Model Applications
- How to choose pre-trained models
- Understanding model complexity, size, latency
- Understanding inference parameters (Temperature, Tokens, Max Length)
3.2 Prompt Engineering
- Prompt types: Zero-shot, Single-shot, Few-shot, Chain-of-thought, Negative prompts
- Prompt best practices (clarity, specificity, guardrails)
- Risks: prompt injection, poisoning, jailbreaking
3.3 RAG (Retrieval Augmented Generation)
- Concept & workflow
- When to use RAG
- AWS tools for RAG:
- Bedrock RAG Knowledge Bases
- Amazon Kendra
- OpenSearch as vector DB
- Aurora / DocumentDB / DynamoDB options
3.4 Fine-tuning Foundation Models
- Pre-training, fine-tuning, continuous tuning
- Instruction tuning
- Data preparation
- RLHF basics
- Cost trade-offs
3.5 Evaluating AI Models
- Human evaluation
- ROUGE
- BLEU
- BERTScore
- Business metrics (revenue, conversion, efficiency)
Module 4: Responsible AI (Week 7)
(Aligned with Domain 4 – 14%)
4.1 Principles of Responsible AI
- Bias, fairness, inclusivity
- Robustness & reliability
- Transparency
- Safety & ethical considerations
4.2 AWS Responsible AI Tools
- Amazon Bedrock Guardrails
- Amazon A2I – Human review
- SageMaker Clarify
- Model Monitor
4.3 Explainable AI
- Transparent vs non-transparent models
- SageMaker Model Cards
- Human-centered design principles
Module 5: Security, Governance & Compliance (Week 8)
(Aligned with Domain 5 – 14%)
5.1 Securing AI Systems
- IAM roles, policies, permissions
- Encryption at rest & in transit
- AWS PrivateLink
- Amazon Macie (PII detection)
- AWS Shared Responsibility Model
5.2 Governance & Compliance
- ISO, SOC compliance
- AI governance frameworks
- AWS services for governance:
- AWS Config
- AWS CloudTrail
- AWS Trusted Advisor
- AWS Audit Manager
- AWS Artifact
5.3 Data Governance
- Data lifecycle
- Data residency
- Logging & monitoring
- Data retention & cataloging
Module 6: Exam Preparation (Last 2–3 Classes)
- Understand AIF-C01 exam format
- Practice MCQs (updated to latest exam pattern)
- Case-study questions
- Matching & ordering questions
- Mock exam #1
- Mock exam #2
- Tips for passing on the first attempt
- How to schedule exam & prepare final checklist
