1. Course Syllabus

   1.  Module 1: Introduction to Dataiku & Data Science Fundamentals – Overview of the Dataiku platform, interface (projects, Flow, datasets), and collaborative analytics. Introduce data science methodology (CRISP-DM) and roles (analyst, data scientist, engineer, business user). Get hands-on by creating a new project, importing a sample dataset, and exploring the Flow; cover basic concepts like managed datasets and versioning.
   2.  Module 2: Getting Data into Dataiku – Data connections and ingestion. Learn to connect to diverse data sources (files, databases, cloud storage). Hands-on: configure connections (e.g. JDBC for SQL databases, Snowflake, Azure Blob) and import data into Dataiku projects. Explore Dataiku’s Data Catalog and dataset metadata.
   3.  Module 3: Data Preparation & Visual Recipes – Data wrangling and cleaning with visual recipes. Use point-and-click recipes (Join, Split, Filter, Group, Prepare, etc.) to transform data without coding. Perform data profiling, remove duplicates/missing values, normalize and encode fields. Hands-on lab: build a data pipeline using visual recipes and review data lineage.
   4.  Module 4: SQL & Database Integration – Leverage SQL within Dataiku. Use SQL Query and Script recipes to write custom transformations. Practice pushdown execution: Dataiku sends queries directly to the database (e.g. Snowflake) for fast in-database processing. Configure a SQL database connection (MySQL/Postgres/Snowflake). Hands-on: write a SELECT query to join tables in-database and output the result as a new dataset.
   5.  Module 5: Python Scripting in Dataiku – Use Python for advanced analysis. Set up Python code environments and use built-in libraries (pandas, NumPy, scikit-learn, XGBoost). Author Python recipes and Jupyter notebooks to manipulate datasets via the Dataiku API. Demonstrate how to call Dataiku functions from Python for data I/O. Hands-on: write a Python recipe to compute custom features or train a scikit-learn model.
   6.  Module 6: R Programming in Dataiku – Use R for statistical analysis. Activate R integration and use R packages (tidyverse, ggplot2, randomForest, etc.). Create R recipes or notebooks for data exploration and modeling. Example: build an R script to perform linear regression or visualize data with ggplot2. Learn how Dataiku handles R environments and packages.
   7.  Module 7: Machine Learning & Automated Modeling – Introduction to Dataiku’s AutoML. Use the Visual ML (Auto Model) interface to build classification/regression models step by step. Compare algorithms (decision trees, logistic regression, XGBoost, etc.) and tune hyperparameters. Explain built-in feature engineering (categorical encoding, feature derivation). Hands-on: solve a predictive problem (e.g. customer churn) using AutoML and review performance metrics.
   8.  Module 8: Model Evaluation & Interpretability – Evaluate and compare models using metrics (accuracy, RMSE, ROC AUC). Use Dataiku’s charts (confusion matrix, lift charts) and explainability tools (feature importance, SHAP). Practice model validation (cross-validation, holdout sets) and understand overfitting. Hands-on: interpret model results, identify key drivers, and prepare a report for stakeholders.
   9.  Module 9: Automation, MLOps & Deployment – Productionize Dataiku projects. Build Scenarios to automate tasks (scheduling, retraining triggers). Create API services from models and deploy webapps or dashboards. Manage model lifecycle: version control, audit logs, monitoring for drift. Cover Dataiku’s governance features. Scenario labs: set up a scheduled model training pipeline and publish a real-time scoring API.
   10.  Module 10: Cloud & Big-Data Integration – Connect Dataiku to cloud platforms and big-data engines. Snowflake: configure a Snowflake data warehouse connection and use pushdown SQL to query large datasets. Azure: leverage Dataiku on Azure (via Cloud Stack) and connect to Azure services (Synapse/Purview). Learn about Azure Synapse integration (DSS supports visual recipes in Synapse). Also cover Hadoop/Spark integration basics for large-scale processing. Hands-on: load data from Azure Blob/Synapse and run a recipe in-database.
   11.  Module 11: Industry Use Cases & Capstone Project – Apply skills to real-world scenarios. Study industry examples (e.g. retail sales forecasting, credit-card fraud detection, manufacturing maintenance). Final capstone: end-to-end project (data ingestion ▶ preparation ▶ modeling ▶ deployment) on a chosen dataset. Prepare for the Dataiku Core Designer certification by reviewing exam topics and taking practice quizzes. Present project results to the class, emphasizing business impact.

Key Features

• Hands-On Labs & Exercises: Each module includes guided labs using real datasets (finance, retail, IoT, etc.) so learners can practice Dataiku recipes, notebooks, and cloud connections in context.
• Certification Prep: Content aligns with Dataiku certification (e.g. Core Designer), including sample questions and review of platform concepts. Learners earn badges or mini-certificates after key assessments.
• Use Cases & Projects: Training leverages industry examples (fraud detection, predictive maintenance, sales forecasting) and a capstone project to illustrate how Dataiku solves business problems. For example, a credit-card fraud detection scenario highlights Dataiku’s unified workflow for data wrangling, feature engineering, and ML.
• Multi-Role Collaboration: Emphasis on Dataiku’s cross-functional platform: business users can use no-code visual tools, while analysts and engineers can write code. Dataiku “provides one solution for all stakeholders… No-, low-, and full-code functionality”, enabling IT and business to work together.
• Integrated Tech Stack: Learners practice with Python, R, and SQL inside Dataiku. The course covers connecting to cloud warehouses (Snowflake with pushdown) and Azure services, giving exposure to modern data platforms and big-data processing.
• Progressive Learning Path: Begins with foundational concepts for beginners and advances to complex topics (MLOps, cloud integration). Each section balances theory (principles of data science, model evaluation) with practical workflows, ensuring a deep, hands-on understanding of Dataiku.