Artificial intelligence methodology

Artificial intelligence (AI) methodology is crucial for successfully applying AI in various domains. It provides a structured approach to data analysis, system design, and model deployment, ensuring that AI solutions are efficient, scalable, and reliable.

Data Analysis for AI

Understanding how to analyze data effectively is foundational for AI. This includes data preparation, visualization, and inference, which are critical for building robust AI models.

AI System Design

A well-structured approach to designing AI systems includes setting clear objectives, iterative development, and ensuring data integrity through proper engineering and processing techniques.

Model Deployment and Monitoring

Deploying AI models involves challenges such as managing data distribution shifts and ensuring continuous learning. Effective monitoring and infrastructure are essential for maintaining model performance in production environments.

Applied AI Methodology

  1. Understanding how analysis works
  2. Using data analysis effectively with machine learning
  3. Determining what AI can achieve

  1. The data science process
  2. Machine learning in business
  3. Data capture
  4. Data preparation
  5. Data visualization
  6. Inference
  7. Data engineering
  8. Rule base approach vs Data science approach

  1. Introduction to AI systems design
    1. Business and ML objectives
    2. Requirements for AI Systems
    3. Iterative process
    4. Framing AI problems
    5. Mind versus data
  2. Data engineering fundamentals
    1. Data sources
    2. Data formats
    3. Data models
    4. Data storage engines and processing
    5. Modes of Dataflow
    6. Batch processing versus stream processing
  3. Training data
    1. Sampling
    2. Labeling
    3. Class imbalance
    4. Data augmentation
  4. Feature engineering
    1. Learned features versus engineered features
    2. Common features Engineering operations
    3. Data Leakage
    4. Engineering good features
  5. Model development and offline evaluation
    1. Model development and trainingç
    2. Model offline evaluation
  6. Model deployment and prediction service
    1. Machine learning deployment Myths
    2. Batch prediction versus online prediction
    3. Model compression
    4. ML on the cloud and on the edge
  7. Data distribution shifts and monitoring
    1. Causes of AI system failures
    2. Data distribution shifts
    3. Monitoring and observability
  8. Continual learning and test in production
    1. Continual learning
    2. Test in Production
  9. Infrastructure and tooling for MLOps
    1. Storage and compute
    2. Development environment
    3. Resource Management
    4. ML Plataform
    5. Build versus buy
  10. The human side of AI
    1. User experience
    2. Team structure
    3. Responsible AI

  1. Introduction to XAI
  2. Defining explanation methods and approaches
  3. Evaluating the quality of explainability methods
  4. Types of model explainability methods
  5. Knowledge extraction methods
  6. Influence-based methods
  7. Example-based methods

  1. AI engineer profile
  2. Important factors
    1. Elucidating the differences between a data scientist and AI engineer
    2. Focusing on simplicity in all project work to reduce risk
    3. Applying agile fundamentals to AI project work
    4. Differences and similarities between DevOps and MLOps
  3. Planning and scoping a project
  4. Communication and logistics of the project
  5. Planning and Researching an AI project
  6. Testing and Evaluating an AI project
  7. Moving from prototype to MVP
Artificial Intelligence methodology

Methods and Tools in AI

Applied AI in Sciences

Develop Artificial Intelligence skills in your field with our free course!

Our course is free and designed to develop skills as an applied AI engineer