Course Outline
Module 1: Essential Python for Machine Learning Workflows
• Programme initiation and environment configuration
Align on project objectives and establish a reproducible Python ML workspace
• Python language fundamentals (accelerated pace)
Review syntax, control structures, functions, and patterns prevalent in ML codebases
• Data structures for machine learning
Utilizing lists, dictionaries, sets, and tuples for features, labels, and metadata
• Comprehensions and functional utilities
Implementing transformations through comprehensions and higher-order functions
• Object-oriented Python for ML developers
Mastering classes, methods, composition, and practical design decisions
• Dataclasses and lightweight modelling
Creating typed containers for configuration, sample data, and results
• Decorators and context managers
Implementing patterns for timing, caching, logging, and resource-safe execution
• File and path management
Handling datasets robustly and managing serialization formats
• Exceptions and defensive programming
Writing ML scripts that fail safely and transparently
• Modules, packages, and project architecture
Structuring reusable ML codebases for maintainability
• Typing and code quality standards
Applying type hints, documentation, and lint-friendly structures
Module 2: Numerical Python, SciPy, and Data Handling
• NumPy foundations for vectorised computing
Executing efficient array operations and performance-conscious coding
• Indexing, slicing, broadcasting, and shapes
Ensuring safe tensor manipulation and shape reasoning
• Linear algebra essentials with NumPy and SciPy
Performing stable matrix operations and decompositions used in ML
• Deep dive into SciPy
Exploring statistics, optimization, curve fitting, and sparse matrices
• Pandas for tabular ML data
Cleaning, joining, aggregating, and preparing datasets effectively
• Deep dive into scikit-learn
Leveraging the estimator interface, pipelines, and reproducible workflows
• Visualization essentials
Creating diagnostic plots for data exploration and model behavior analysis
Module 3: Programming Patterns for Building ML Applications
• Transitioning from notebooks to maintainable projects
Refactoring exploratory code into structured, reusable packages
• Configuration management
Managing externalized parameters and implementing startup validation
• Logging, warnings, and observability
Implementing structured logging for debuggable ML systems
• Building reusable components with OOP and composition
Designing extensible transformers and predictors
• Practical design patterns
Applying Pipeline, Factory/Registry, Strategy, and Adapter patterns
• Data validation and schema checks
Preventing silent data issues before they impact models
• Performance optimization and profiling
Identifying bottlenecks and applying effective optimization techniques
• Model I/O and inference interfaces
Ensuring safe data persistence and clean prediction interfaces
• End-to-end mini project
Constructing a production-style ML pipeline with configuration and logging
Module 4: Statistical Learning for Tabular, Text, and Image Data
• Evaluation foundations
Mastering train/validation splits, honest cross-validation, and business-aligned metrics
• Advanced tabular machine learning
Utilizing regularized GLMs, tree ensembles, and leakage-free preprocessing
• Calibration and uncertainty estimation
Applying Platt scaling, isotonic regression, bootstrap, and conformal prediction
• Classical NLP methods
Understanding tokenization trade-offs, TF-IDF, linear models, and Naive Bayes
• Topic modelling
Grasping LDA fundamentals and understanding practical limitations
• Classical computer vision techniques
Implementing HOG, PCA, and feature-based pipelines
• Error analysis
Detecting bias, label noise, and spurious correlations
• Hands-on labs
Building a leakage-proof tabular pipeline
Comparing and interpreting text baselines
Conducting structured failure analysis for classical vision baselines
Module 5: Neural Networks for Tabular, Text, and Image Data
• Mastering the training loop
Implementing clean PyTorch loops with AMP, clipping, and reproducibility measures
• Optimization and regularization techniques
Managing initialization, normalization, optimizers, and schedulers
• Mixed precision and scaling strategies
Implementing gradient accumulation and checkpointing strategies
• Neural networks for tabular data
Utilizing categorical embeddings, feature crosses, and ablation studies
• Neural networks for text data
Working with embeddings, CNNs, BiLSTMs/GRUs, and sequence handling
• Neural networks for vision data
Understanding CNN fundamentals and ResNet-style architectures
• Hands-on labs
Developing a reusable training framework
Comparing Tabular NN vs. boosting models
Experimenting with CNN augmentation and scheduling
Module 6: Advanced Neural Architectures
• Transfer learning strategies
Implementing freeze/unfreeze patterns and discriminative learning rates
• Transformer architectures for text
Understanding self-attention internals and fine-tuning approaches
• Vision backbones and dense prediction
Exploring ResNet, EfficientNet, Vision Transformers, and U-Net concepts
• Advanced architectures for tabular data
Utilizing TabTransformer, FT-Transformer, and Deep & Cross networks
• Time series considerations
Managing temporal splits and detecting covariate shift
• PEFT and efficiency techniques
Evaluating LoRA, distillation, and quantization trade-offs
• Hands-on labs
Fine-tuning pretrained text transformers
Fine-tuning pretrained vision models
Comparing Tabular transformers vs. GBDT models
Module 7: Generative AI Systems
• Fundamentals of prompting
Executing structured prompting and controlled generation techniques
• LLM foundations
Understanding tokenization, instruction tuning, and hallucination mitigation
• Retrieval-Augmented Generation (RAG)
Implementing chunking, embeddings, hybrid search, and evaluation metrics
• Fine-tuning strategies
Applying LoRA and QLoRA with strict data quality controls
• Diffusion models
Grasping latent diffusion intuition and practical adaptation methods
• Synthetic tabular data generation
Using CTGAN and addressing privacy considerations
• Hands-on labs
Building a production-style RAG mini-application
Validating structured output with schema enforcement
Optional diffusion experimentation
Module 8: AI Agents and MCP
• Agent loop design
Implementing observe, plan, act, reflect, and persist cycles
• Agent architectures
Exploring ReAct, plan-and-execute, and multi-agent coordination strategies
• Memory management
Utilizing episodic, semantic, and scratchpad memory approaches
• Tool integration and safety
Defining tool contracts, implementing sandboxing, and defending against prompt injection
• Evaluation frameworks
Using replayable traces, task suites, and regression testing
• MCP and protocol-based interoperability
Designing MCP servers with secure tool exposure mechanisms
• Hands-on labs
Building an agent from scratch
Exposing tools via an MCP-style server
Creating an evaluation harness with safety constraints
Requirements
Participants are expected to possess a practical understanding of Python programming.
This programme is designed for technical professionals with intermediate to advanced skill levels.
Testimonials (2)
the ML ecosystem not only MLFlow but Optuna, hyperops, docker , docker-compose
Guillaume GAUTIER - OLEA MEDICAL
Course - MLflow
I enjoyed participating in the Kubeflow training, which was held remotely. This training allowed me to consolidate my knowledge for AWS services, K8s, all the devOps tools around Kubeflow which are the necessary bases to properly tackle the subject. I wanted to thank Malawski Marcin for his patience and professionalism for training and advice on best practices. Malawski approaches the subject from different angles, different deployment tools Ansible, EKS kubectl, Terraform. Now I am definitely convinced that I am going into the right field of application.
