UiPath AI Center Intelligent Automation Training Course

Course overview

UiPath automation teams often reach the point where rules-based workflows cannot reliably classify emails, extract variable document fields, interpret customer intent or make predictions from operational data. AI Center provides the production bridge between machine-learning models and UiPath robots, but teams need more than a model upload: they need governed deployment, repeatable testing, human validation, monitoring and a clear operating model. This course addresses the practical gap between building an attended or unattended automation and running AI-enabled processes safely at enterprise scale.

Participants learn to use UiPath AI Center to provision and manage AI skills, deploy packaged models, configure model versions and connect predictions to UiPath Studio workflows. The course covers the AI Center project structure, ML packages, datasets, pipelines, endpoints, predictions, logging and model lifecycle controls. Participants also build Document Understanding workflows that combine OCR, classification, extraction and Validation Station review, then examine how Orchestrator queues, robot execution and AI Center endpoints work together in a resilient business process.

Delivery is instructor-led and workshop-based, using a realistic invoice and correspondence-processing case. Participants configure an AI Center project, deploy and test an ML skill, call it from Studio, route low-confidence results for human review and prepare operational monitoring measures. They leave with a working AI-enabled automation solution pack: a Studio workflow, AI Center deployment configuration, test evidence, exception-handling design and an implementation plan suitable for adapting to their own use case.

The programme suits experienced UiPath developers, automation leads, solution architects and data practitioners who need to move from proof-of-concept AI to supportable intelligent automation.

Course objectives

By the end of this course, participants will be able to:

• Configure UiPath AI Center projects, permissions and compute settings for an enterprise automation use case
• Package, deploy and version an ML model as an AI Center ML skill
• Create and manage datasets, labels and training data workflows for document-based AI use cases
• Build UiPath Studio workflows that invoke AI Center prediction endpoints and process model outputs
• Design confidence-score thresholds, retry logic and human-in-the-loop validation routes
• Implement Document Understanding classification, extraction and Validation Station review workflows
• Test AI-enabled automations using representative cases, acceptance criteria and exception scenarios
• Produce an AI Center deployment and monitoring plan covering model versions, logs, ownership and rollback

Benefits of attending

Target competencies

Who should attend

• UiPath Developers — who need to embed deployed ML skills and document AI into production robot workflows
• Intelligent Automation Leads — who must select, govern and scale AI use cases across an automation portfolio
• RPA Solution Architects — who design the integration, security and lifecycle architecture for AI-enabled processes
• Data Scientists — who need to operationalise models through AI Center for consumption by UiPath robots
• Document Processing Specialists — who build extraction workflows for invoices, claims, forms and correspondence
• Automation Operations Managers — who oversee robot reliability, exception handling and model performance in production

Requirements and prerequisites

Participants should have working experience building and debugging UiPath Studio automations, including variables, arguments, control flow, selectors, activities and exception handling. Familiarity with UiPath Orchestrator concepts such as folders, processes, robots, assets and queues is expected. Participants should also understand basic machine-learning terms, including training data, inference, model version, confidence score and false positive; they do not need to build models in Python. Prior data-science experience, advanced mathematics and previous AI Center administration are not required. Access to a UiPath environment with AI Center and Document Understanding enabled is strongly recommended for hands-on practice.

Training methodology

The instructor demonstrates each AI Center capability in a live UiPath environment, then participants apply it in guided build sessions. Short technical briefings explain the model lifecycle, endpoint behaviour and governance decisions behind each activity. Teams work through an invoice and customer-correspondence case, configuring an ML skill, integrating it with Studio and resolving low-confidence exceptions through Document Understanding validation. Daily reviews compare implementation choices, test results and operational risks. The final workshop converts the completed prototype into a deployment, monitoring and ownership plan for a real workplace candidate process.

Course outline

Day 1: AI Center architecture and intelligent automation design

• UiPath AI Center architecture, services and tenant-level dependencies
• AI Center projects, access roles and separation of development and production environments
• ML packages, ML skills, endpoints and prediction requests
• Model lifecycle stages from data preparation to production inference
• Intelligent automation use-case assessment using volume, variability, data quality and risk
• Confidence scores, thresholds and the economics of human review
• AI governance foundations for explainability, ownership, auditability and rollback

Workshop: Participants assess a document-processing scenario and produce an AI Center solution canvas with process scope, data inputs, model decision points and control requirements.

Day 2: Deploying and managing AI Center ML skills

• AI Center project creation and resource configuration
• ML package formats, package metadata and deployment prerequisites
• Deploying an ML package as a managed ML skill
• ML skill configuration, endpoint access and version management
• Datasets, data labelling concepts and training-data quality checks
• Pipelines for training, evaluation and model deployment
• Prediction logging, endpoint testing and initial performance validation

Workshop: Participants create an AI Center project, deploy a supplied document-classification ML package and record endpoint tests in a model deployment checklist.

Day 3: Integrating AI Center predictions with UiPath Studio

• UiPath Studio project structure for AI-enabled automations
• Calling AI Center ML skills through prediction activities and service endpoints
• Mapping JSON prediction outputs to workflow variables and business fields
• Input preparation for files, text, images and structured records
• Confidence-based branching and business-rule reconciliation
• Exception handling for unavailable endpoints, invalid payloads and low-confidence predictions
• Orchestrator queues and transaction design for scalable AI-assisted processing

Workshop: Participants build a Studio workflow that submits documents to an ML skill, interprets prediction results and routes exceptions through a queue-based process.

Day 4: Document Understanding and human-in-the-loop operations

• UiPath Document Understanding architecture and processing stages
• Digitisation using OCR engines and document quality controls
• Document classification methods and classifier selection
• Extraction taxonomy, field definitions and extractor configuration
• Combining ML extraction with rule-based validation and business logic
• Validation Station configuration and action-centre human review patterns
• Measuring extraction accuracy, validation effort and straight-through-processing rates

Workshop: Participants configure an invoice-processing workflow with classification, extraction, confidence rules and Validation Station review, producing a tested sample batch.

Day 5: Production readiness, monitoring and implementation planning

• Acceptance criteria for AI-enabled automation and model performance
• Test design using representative documents, edge cases and negative scenarios
• Production monitoring for prediction failures, confidence trends and exception volumes
• Model version control, release approvals and rollback procedures
• Security and data-handling controls for documents, model inputs and predictions
• Operating-model roles across automation, data science, business operations and support
• Use-case roadmap development and benefits tracking for intelligent automation

Workshop: Participants complete an end-to-end capstone and produce a deployment runbook, monitoring dashboard specification and 90-day implementation plan for their selected use case.

Tools & standards covered

UiPath AI Center, UiPath Studio, UiPath Orchestrator, UiPath Document Understanding

A typical training day

Live online deliveries follow the same structure in the East Africa Time zone, with shorter screen blocks and longer breaks.

What the fee includes

How you can take this course

Certification

Participants who complete the full five days receive the Skillset Development Certificate of Completion, stating the course title, course code, dates and delivery format — suitable for professional-development records and employer reimbursement.

Frequently asked questions