Plan du cours
Introduction to AI Builder and Low-Code AI
- AI Builder capabilities and common scenarios
- Licensing, governance, and tenant-level considerations
- Overview of the Power Platform integrations (Power Apps, Power Automate, Dataverse)
OCR and Form Processing: Structured and Unstructured Documents
- Differences between structured templates and free-form documents
- Preparing training data: labeling fields, sample diversity, and quality guidelines
- Building an AI Builder form processing model and evaluating extraction accuracy
- Post-processing extracted data: validation, normalization, and error handling
- Hands-on lab: OCR extraction from mixed form types and integration into a processing flow
Prediction Models: Classification and Regression
- Problem framing: qualitative (classification) vs quantitative (regression) tasks
- Feature preparation and handling missing data within Power Platform workflows
- Training, testing, and interpreting model metrics (accuracy, precision, recall, RMSE)
- Model explainability and fairness considerations in business use cases
- Hands-on lab: build a custom prediction model for churn/score or numeric forecast
Integration with Power Apps and Power Automate
- Embedding AI Builder models into canvas and model-driven apps
- Creating automated flows to process extracted data and trigger business actions
- Design patterns for scalable, maintainable AI-driven apps
- Hands-on lab: end-to-end scenario — document upload, OCR, prediction, and workflow automation
Complementary Process Mining Concepts (Optional)
- How Process Mining helps discover, analyze and improve processes using event logs
- Using Process Mining outputs to inform model features and automate improvement loops
- Practical example: combine Process Mining insights with AI Builder to reduce manual exceptions
Production Considerations, Governance, and Monitoring
- Data governance, privacy, and compliance when using AI Builder on sensitive documents
- Model lifecycle: retraining, versioning, and performance monitoring
- Operationalizing models with alerts, dashboards, and human-in-the-loop validation
Summary and Next Steps
Pré requis
- Experience with Power Apps, Power Automate, or Power Platform administration
- Familiarity with data concepts, basic ML ideas, and model evaluation
- Comfort working with datasets, Excel/CSV exports, and basic data cleansing
Audience
- Power Platform developers and solution architects
- Data analysts and process owners seeking automation through AI
- Business automation leads focused on document processing and prediction use cases
Nos clients témoignent (2)
J'ai trouvé le formateur très engageant et très rapide pour répondre aux questions liées à notre travail. Il a vraiment adapté l'enseignement à nos besoins et s'est surpassé pour les satisfaire. Je ne saurais trop recommander Shaun !
Tom King - Complete Coherence
Formation - Microsoft Power Platform Fundamentals
Traduction automatique
Je suis vraiment admiratif de la patience du formateur face à toutes les personnes qui lui demandaient de répéter quelque chose 4-5 fois. Je pense également qu'il possède une grande expertise sur le sujet, mais comme mentionné plus haut, nous n'avons pas passé assez de temps dessus. De plus, il était bon que la formation soit pratique, où nous pouvions mettre en pratique en temps réel ce qui nous était enseigné. Cependant, je voudrais en savoir plus sur les PowerApps et moins sur SharePoint, car je suis déjà très familier avec celui-ci. Si j'avais voulu en apprendre davantage, je choisirais probablement une formation sur le SharePoint plutôt que sur les PowerApps.
Patrycja - EY GDS
Formation - Microsoft Flow/Power Automate
Traduction automatique