STELA Advanced RPA develops your ability to operate a business automation when a response is lost, a process stops midway, or another person must take over. Build a robot that records what is known, makes uncertainty visible, and supports a justified decision about the next action. The course uses fictional requests, STELA STORE, and the spreadsheet and automation capabilities introduced in Fundamentals. Design the process around observable states and clear conditions for entering and leaving each stage. Practice recovery from failures before and after a business effect, and verify persisted results before submitting an operation again. Define bounded retries, prevent duplicates, and reconcile individual records as well as batch totals. Human intervention includes an owner, evidence, a recorded decision, and a condition for resuming work. Extend your work with OCR and IA GPT by separating extraction, interpretation, and authorization. Evaluate different inputs, preserve missing or ambiguous information, and compare behavior before and after a change. These activities build your ability to assess an automation using evidence rather than the apparent confidence of a response. The final project brings recovery, reconciliation, testing, business acceptance, and operational handover together. Another person must be able to use the guide, handle an exception, and restore a compatible earlier robot version. Practical STELA RPA Fundamentals skills and access to the assigned lab are required. No programming is needed; the course concentrates on recovery and operation using the technologies already introduced.

– Getting started: demonstrate the Fundamentals entry competencies and prepare the workbook, data, and tracking materials.
– State-based design: define observable states, stage contracts, and evidence that supports each transition.
– Recovery: identify failure scope, bound retries, and review recovery proposals from Robot Maker and Computer Vision.
– Resuming without duplicates: query persisted effects and test interruptions before and after a business action.
– Process consistency: reconcile records and totals, preserve partial results, and control compensation and uncertain outcomes.
– IA GPT and unstructured data: validate OCR and interpretation, evaluate an input corpus, and compare versions with regression checks.
– Operation and maintenance: deliver identifiable versions, obtain business acceptance, test the operating guide, and restore compatible versions.
– Final project: demonstrate failures and recovery, work with a second operator, and submit a reviewable handover package.