Thursday, 28 May 2026

Interview Questions(Fusion and OIC) 2026-May

 Interview Questions(Fusion and OIC) 



1. Can you please brief me about your Oracle Fusion and OIC application career?

Answer:

Over my XX-year career in enterprise application integration, the last several years have been dedicated exclusively to the Oracle Cloud Ecosystem, specializing as a Lead Architect across Oracle Fusion Cloud (ERP, HCM, Procurement) and Oracle Integration Cloud (OIC).

My expertise spans designing real-time, event-driven integrations using REST/SOAP frameworks, orchestrating complex bulk data migrations using FBDI and BIP data extracts, and decoupling heavy legacy PL/SQL business logic from ATP databases into OIC cloud-native design patterns. 

I frequently act as the bridge between technical delivery teams and functional stakeholders, establishing end-to-end integration roadmaps, security strategies, and performance tuning for multi-tier global rollouts.

-------------------------------

2. When you will use REST and SOAP web services?

Answer:

REST APIs: 

I leverage REST for high-performance, synchronous, lightweight, and stateless record-level operations (CRUD). Because it utilizes JSON payloads, it consumes significantly less bandwidth and provides fast execution, making it the strategic choice for modern web applications, mobile interfaces, and real-time point-to-point updates (Ex:-Triggering an immediate customer update from a CRM).

SOAP Web Services: 

I default to SOAP when dealing with complex business process orchestrations or deep, heavily nested parent-child hierarchies that must be committed under strict transactional boundaries (WS-Security / ACID compliance). 

It is also essential when interfacing with legacy middleware platforms or enterprise service buses (ESBs) that strictly require formal, contract-driven WSDL agreements.

----------------------

3. Suppose, Oracle has both REST and SOAP web services available (ex: creation of Supplier or creation of Invoices). Apart from this we can also use FBDI process. So, which one you will use it on what scenario and why?

Answer:

My architectural choice is driven by Volume, Frequency, and Data Complexity:

REST API: 

I will choose REST if the source system sends single transactions or micro-batches in real time (e.g., an e-commerce platform pushing an invoice immediately upon checkout). Why: It offers instant synchronous feedback, lightweight JSON processing, and immediate error handling.

-------------------------

4. When will you go for the FBDI approach?

Answer:

I enforce the FBDI approach under two distinct scenarios:

Data Conversion/Migration Phases: Moving historical open balances, 20,000+ supplier master records, or active purchase orders from a legacy ERP system to Oracle Fusion during initial implementation go-lives.

High-Volume Inbound Batch Interfaces:

Daily or weekly scheduled bulk processing. For instance, if an external global procurement or specialized inventory system generates a daily CSV extract containing 5,000+ receiving transactions or AP invoices to be synchronized into Fusion.

SOAP Service: 

I will choose SOAP if the REST endpoint lacks functional schema parity for a highly specific sub-entity (e.g., complex banking profile assignments nested deep within a supplier profile) or if a legacy middleware system cannot parse JSON.

FBDI Process: 

I will strictly use FBDI if the data volume exceeds record-level thresholds (e.g., thousands of invoices pushed as a nightly batch file). Why: Web services will hit HTTP timeout limits or rate-limiting thresholds (typically around 5,000 calls per hour) when processing massive volumes. FBDI handles bulk imports asynchronously at the database layer via staging tables and ESS jobs.

-------------------

5. I have developed an integration and wanted to make it more secure. What are the different things that you will do to secure the integration? Tell me the answer with a real-time example.

Answer:

I apply a defense-in-depth security strategy spanning multiple layers.

Real-Time Example: 

Consuming a sensitive payroll and direct-deposit integration between an external HR system (like ADP) and Oracle HCM Cloud.

Security Actions Taken:

Network Layer: 

Configure IP Whitelisting / Allowlisting on the OIC API Gateway to drop traffic from any source outside ADP's explicit public IP blocks.

Authentication Layer: 

Decommission basic authentication and implement OAuth 2.0 Client Credentials Grant utilizing short-lived access tokens.

Authorization Layer: 

Apply the principle of least privilege; the OIC integration service user in Fusion is mapped to a custom role restricting access only to the necessary employee/bank accounts REST resources.

Payload Security: 

Enforce PGP Encryption for files sitting on intermediate staging SFTP locations.

Data Masking: 

Turn off payload tracking in production and programmatically redact/mask high-risk fields (such as SSNs or Bank Account Numbers) within the OIC error-handling logs to prevent data leaks.

--------------------

6. Explain about OAuth2.0?

Answer:

OAuth 2.0 is an industry-standard token-based authorization framework. It allows an external application to gain limited access to Oracle Fusion or OIC resources without ever exposing or hardcoding primary user passwords.

In a standard system-to-system integration, OIC uses the Client Credentials Grant type. The external client sends its secure Client ID and Client Secret to the Identity Provider (like Oracle IDCS) to retrieve a short-lived Access Token (formatted as a JSON Web Token - JWT). This token acts as a digital key valid for a limited window (typically 60 minutes) and contains restricted scopes (permissions). The client then attaches this string to the HTTP header as a Bearer Token to invoke the target APIs securely.

--------------------

7. Do you have any experience with VBCS?

Answer:

Yes, I have extensive experience leveraging Visual Builder Cloud Service (VBCS) to create custom UI extensions and composite applications that fill functional gaps in Oracle Fusion. I specialize in configuring VBCS Service Connections to consume Fusion REST APIs natively, managing business objects, mapping data flows, and implementing role-based component security directly within the custom user interfaces to enhance the overall user experience.

--------------------

8. What is the stage file operation max size?

Answer:

In Oracle Integration Cloud (OIC), the maximum size file that can be processed by the Stage File action in-memory is 50 MB.

For handling larger data files (up to 1 GB), we must utilize the Read File in Chunks option within the Stage File configuration. This streams the data sequentially using a looping background mechanism to prevent JVM out-of-memory exceptions.

-----------------------

9. Any challenges with FTP/SFTP or ATP DB adapters or Salesforce Adapter?

Answer:

Yes, throughout my delivery experience, I have managed and resolved several specific adapter limitations:

FTP/SFTP Adapter: 

Faced connection timeout issues and file locking constraints when large files were being written to concurrently. Resolved this by introducing strict file-naming conventions (.tmp shifting to .csv post-write), optimizing pool settings, and configuring explicit retry limits on the OIC connection properties.

ATP DB Adapter: 

Encountered transaction timeout thresholds during long-running PL/SQL package executions. I resolved this by decoupling heavy transactional logic, moving data processing out of database-level packages, and converting the long-running database steps into modular, asynchronous OIC SQL Lookups, or breaking the data into smaller chunks.


Salesforce Adapter: 

Dealt with schema drift issues and handling bulk API limits when processing millions of change-event records. Solved this by setting up proper custom fields mapping, relying on Salesforce's daily API limits monitoring, and structuring batch sizes effectively within the integration mapping layers.

--------------

10. I have created a scheduled integration to create invoices and want to convert it into an app-driven integration. How can we do that?

Answer: 

Since OIC does not allow a direct structural change to the integration type pattern on an existing canvas, the migration requires a precise rebuild strategy:

Export Mappings: Open the current Scheduled Integration, navigate to the main data mapping transformation node for the Invoice payload, and Export the underlying .xsl map file locally.

Create App-Driven Shell: Create a brand new integration selecting the App-Driven Orchestration design pattern.

Configure REST Trigger: Drag a REST adapter to the initial Trigger node. Configure a POST operation with a relative URI (e.g., /createInvoice), and supply a sample JSON payload representing the incoming invoice request format.

Re-create Invoke & Import Maps: Add your target Oracle ERP Cloud Invoice creation adapter node. Open the newly generated Map node, click Import, and upload the .xsl file saved in Step 1 to instantly restore the data transformation logic.

Clean up Staging Elements: If the scheduled flow used a Stage File loop to parse a bulk file, remove that structural loop since the app-driven trigger now accepts the individual transaction payloads directly in real time.

-----------------------

11. How will you identify which subject areas need to be used?

Answer:

To accurately identify the correct OTBI (Oracle Transactional Business Intelligence) Subject Area, I follow a systematic process:

Analyze the Functional Requirements: 

Map out the target transactional entities needed (eg:- Real-time Invoice validation requires Payables, while asset tracking requires Fixed Assets).

Cross-Reference Tables to Subject Areas: 

I utilize the official Oracle Cloud Subject Area to Database Table Mapping Documentation.Inspect via UI: Navigate to the specific functional page in Fusion, use the "Inspect" tool or the Help --> About This Page feature to identify the underlying View Object (VO).

OTBI Subject Area Guide: Validate the corresponding Subject Area using the Oracle BI Composer or the standard catalog tree structure (eg:- Payables Invoices - Transactions Real Time).

----------------------

12. How will you get DFF attributes in OTBI reports?

Answer:

DFF (Descriptive Flexfield) attributes do not appear in OTBI reports automatically; they must be explicitly synchronized:

Run Deployment: Ensure the DFF is properly configured and successfully deployed within the Manage Descriptive Flexfields setup task in Fusion.

BI Enabled Checkbox: Ensure that the specific DFF segments have the "BI Enabled" checkbox checked within their segment definition settings.

Run Synchronization Job: Submit the enterprise scheduled process: "Import Oracle Fusion System Management Extensibility Applications Data into BI" (commonly referred to as the BIP/OTBI Sync job). Once completed, the attributes materialize under a dedicated sub-folder labeled "Flexfields" within their respective OTBI Subject Area folder hierarchy.

--------------

13. Have you created custom roles in Oracle Fusion?

Answer:

Yes. In accordance with strict internal security compliance, we avoid assigning seeded Oracle roles directly to users. I navigate to the Security Console, find the closest standard Oracle seeded role (such as Accounts Payable Manager), and perform a Deep Copy. From there, I edit the custom role prefixing it (e.g., _AP_MANAGER) to add or remove specific functional security policies or data security policies according to business requirements.

------------------

14. Explain when we will create a data role and when we will create a job role?

Answer:

(Note: In modern Oracle Cloud releases, explicit "Data Roles" have been unified into the Role Common Model, but the structural concept remains critical).

Job Role: Created to define a business function or job description (e.g., General Accountant or Procurement Agent). It acts as a logical container for Duty Roles, specifying what actions (functions, pages, tasks, privileges) a user holding that role can execute across the system.

Data Role (Security Context Assignment): Created or assigned to control which specific data records that user can perform those job actions on. It bridges the Job Role to a concrete security context, such as a specific Business Unit (BU), Ledger, or Inventory Organization. For example, a user has the Job Role of "Accounts Payable Clerk," but their Data Context limits them to processing transactions only within the "US Business Unit.

-------------------

15 Tell some of the Supplier Manager Privileges?

Answer:

Key privileges rolled into a Supplier Manager security profile include:

PO_MANAGE_SUPPLIER_REGISTRATION_PRIV (Manage Supplier Registration)

PO_EDIT_SUPPLIER_PRIV (Edit Supplier profile details, sites, contacts)

PO_VIEW_SUPPLIER_PRIV (View Supplier profile pages)

PO_MANAGE_SUPPLIER_PROFILES_PRIV (Maintain general supplier profile structures)

---------------

16. Can you tell me any one of the complex reports that you have built?

Answer:

I designed a global Cross-Module Reconciliation Report combining data across Payables (AP), Purchasing (PO), and General Ledger (GL) using BI Publisher (BIP).

Complexity: It required stitching complex multi-table joins across AP_INVOICES_ALL, AP_INVOICE_LINES_ALL, PO_HEADERS_ALL, and GL_JE_LINES to trace mismatched multi-currency variances.


Performance Optimization: Because the report targeted millions of records across multi-national ledgers, standard OTBI timed out. I wrote optimized SQL queries using explicit hints, drove filtering criteria efficiently via standard bind parameters, structured the data model as an optimized parent-child XML data dataset, and created a highly formatted Excel template layout tailored for corporate finance auditors.

-------------------

17. Have you done the callback integration?

Answer:

Yes. Callback integrations are standard in my bulk integration designs involving asynchronous file exports or imports. For example, when invoking the ERP Integration Service to run an FBDI file upload, the process is completely asynchronous. Instead of forcing OIC to resource-intensively loop and poll the Oracle Fusion ESS job status every few minutes, I configure the call payload with a Callback URL. Once the Fusion bulk import job completes, Oracle Fusion natively invokes the OIC callback endpoint, passing the final execution status (SUCCESS/ERROR) along with the log summary files for automated processing.

--------------------------

18. Any extension or customization you did?

Answer:

I have delivered multiple functional extensions. A primary example involved extending the Oracle Fusion Procurement cloud page where users needed to run real-time credit checks on high-value vendors. Since this logic didn't exist natively, I configured a custom action button on the Fusion Page using Page Composer, which triggered a secure external REST service call managed by OIC to pull data from a third-party credit rating agency and present the metric on-screen inside a custom descriptive flexfield (DFF).

--------------------------

19. I want to hide a column. I have 5 columns total: For US client show only 3 DFF attributes. For Europe Client show 2 DFF attributes only. How to achieve this task?

Answer: 

This conditional visibility requirement is achieved via Page Composer utilizing EL (Expression Language) Expressions:

Open the target transactional screen in Oracle Fusion and initiate a Sandbox containing the Page Composer tool.

Select the specific DFF column component you want to hide/show and open its Component Properties.

Locate the "Rendered" or "Visible" attribute property box, click the dropdown, and select Expression Builder.

Write an EL Expression that dynamically reads the current user’s security profile context or the session Business Unit attribute.

Example EL Expression: #{bindings.BusinessUnit.inputValue == 'US_BU'}

Apply the inverse logic expression to the European columns. This ensures that when the page renders, the framework checks the runtime transaction context and visually displays or completely hides the respective DFF columns on the client browser dynamically.

-------------------

20. Have you created any drill-down reports for dashboards?

Answer:

Yes, I routinely build interactive analytical dashboards in OTBI. I configure drill-down interactions by building a high-level summary report (e.g., Total Spend by Business Unit chart) and a secondary, granular detailed report (Line-Level Invoice Details). Inside the summary report’s column properties, under the Interaction tab, I define the action link to "Navigate to BI Content" pointing directly to the detailed report. I pass the shared context columns (like Business Unit ID or Fiscal Year) as dynamic prompt parameters so that clicking an element on the dashboard instantly refines and expands into the deep-dive line reports.

-----------------

21. I wanted to see sales order reports or finance reports based on the role?

Answer:

This is managed securely at the BI Catalog Security Folder Level:

In the BI catalog (/shared/Custom), structure your folders by functional domain (e.g., create a /Sales Reports folder and a /Finance Reports folder).

Click on the Permissions setting for the specific folder.

Strip out general public access and add explicit Application Roles matching your user profiles (e.g., assign access to the /Sales Reports folder exclusively to the Customer Service Representative application role, and assign /Finance Reports strictly to the Financial Analyst application role).

This ensures that when a user logs into the BI reporting environment, they are granted a clean, secure view containing only the folders and reporting dashboards mapped to their authorized organizational functions.

------------------

22. Any tasks comes to you. How will you estimate tasks?

Answer:

As a Lead Technical Consultant, I enforce a standardized Estimation Framework based on complexity tiers rather than guessing:

Requirement Discovery & Scoping: Breaking down the target object into one of three complexity tiers:

Simple: Straightforward point-to-point pass-through API with minimal data transformation mapping. (Estimated: 3-5 days).

Medium: Multiple endpoints, conditional data routing, or standard DFF mappings requiring custom lookups. (Estimated: 8-12 days).

Complex: High-volume FBDI pipelines involving complex file chunking, custom error-handling callbacks, or heavy BIP data transformations. (Estimated: XX-25+ days).

Buffer Allocation: Every single estimate accounts for standard lifecycle phases: Design & Mapping (25%), Development & Unit Testing (40%), Functional / UAT Defect Support (20%), and Deployment Preparation (15%).

------------------------------

23. What are the modules that you have worked on?

Answer:

Throughout my integration portfolio, I have heavily supported across three primary Oracle Fusion Cloud modules:

Oracle Financials (ERP): Payables (AP Invoices, Payments), Receivables (AR Customers, Billing), General Ledger (Journal Imports).

Oracle Procurement: Purchasing (PO Creation), Supplier Model (Supplier Profiles, Sites, Bank Accounts).

Oracle Human Capital Management (HCM): Global Human Resources (Worker Onboarding, Assignments), Payroll Interfaces (Extracts, Bank Account updates).

---------------------

24. How will make Functional team properly communicate with Technical team if we hire as Lead Technical Role? 

Answer:

With XX years of industry experience, I have observed that communication breakdowns happen because Functional teams speak in business processes (eg:-'Three-way matching failed'), while Technical teams speak in schemas and code (e.g.,'ORA-00904 invalid identifier').To bridge this gap effectively as a Technical Lead, I will implement a structured integration delivery process:

Mandate Functional Design Documents (FDD) or (FSD): 

I will require the functional team to deliver a formal FDD outlining the complete business logic, source/target field names, and crisp error validation conditions before code construction begins.

Enforce Technical Design Document (TDD or TSD) Walkthroughs: 

I will lead collaborative mapping workshops where our technical developers present their integration mapping logic back to the functional consultants using plain business language to validate rules early.

Standardize an Integration Mapping Matrix: Establish a shared spreadsheet defining clear Source Field --> Transformation Rule --> Target Oracle API Attribute columns. This serves as the single source of truth that aligns both teams, eliminating assumption errors and significantly reducing cycle times during User Acceptance Testing (UAT).



No comments: