Mock IBM i Developer Interview

What This Lesson Prepares You For

This is the capstone lesson for the Interview and Professional Readiness lessons: a single, structured mock interview combining definition questions, code-reading questions, and scenario-based questions, the same mix a real beginner-level IBM i interview is likely to use.

Concepts to Revise First

This lesson assumes you have already worked through, or are ready to review: RPGLE Basic Interview Questions, RPGLE File I/O Interview Scenarios, CLLE Interview Scenarios, SQLRPGLE Interview Scenarios, Display File and Subfile Interview Questions, and Debugging Scenario- Based Questions.

Common Interview or Job Discussion Scenarios

A realistic beginner-level interview mixes three question types in one conversation: quick definitions, a short piece of code to read, and a "what would you do" scenario. This lesson walks through one example of each in sequence, the way they might actually be asked back to back.

Strong Beginner-Level Answer Approach

Across all three question types, the same habit matters most: connecting the answer to something concrete, a real behavior, a real line of code, or a real check, rather than a memorized phrase. The sections below give one worked example of each type, with a strong answer and the reasoning behind it.

Weak Answers to Avoid

• Answering a definition question correctly but being unable to connect it to a code-reading or scenario question later in the same conversation.
• Treating each question as fully separate, instead of noticing when a later question builds on an earlier one.
• Rushing to answer before actually reading a code snippet fully.

Scenario-Based Practice

Definition question: "What is the difference between a physical file and a logical file?"

Strong answer: "A physical file actually stores the data. A logical file provides a different view over that same data, like a different key order or a subset of fields, without duplicating the data itself."

Code-reading question: "What does this snippet do, and is anything missing?"

dcl-f CUSTMAST disk keyed;

dcl-s custNbr packed(6:0);

custNbr = 9999;
chain custNbr CUSTMAST;
dsply CUSTBAL;

Strong answer: "This looks up customer 9999 by key and displays their balance. It's missing a %found(CUSTMAST) check before the dsply. If custNbr doesn't exist in CUSTMAST, CUSTBAL would hold whatever value was left over from before the CHAIN, not a genuine value for this customer, which could silently display misleading data instead of failing outright."

Scenario question: "A batch report job shows as completed in WRKSBMJOB, but no spool file was produced. What would you check?"

Strong answer: "I'd check that specific batch job's own job log first, separate from whatever wrapper submitted it, since a job can end without producing output if it failed partway through after being submitted successfully, exactly the gap investigated in the Batch Job Failure Investigation mini project."

How to Explain Your Thinking Clearly

Notice that all three strong answers above share a structure: state what the thing is or does first, then name the specific detail, check, or mechanism that matters. This structure works whether the question is a definition, a code snippet, or a scenario, which is exactly why it's worth practicing as one repeatable habit rather than three separate skills.

Practical IBM i Example

The corrected version of the code-reading snippet above:

dcl-f CUSTMAST disk keyed;

dcl-s custNbr packed(6:0);

custNbr = 9999;
chain custNbr CUSTMAST;
if %found(CUSTMAST);
  dsply CUSTBAL;
else;
  dsply 'Customer not found';
endif;

Being able to produce this corrected version, not just identify the problem, is a strong follow-up to have ready.

Mini Practice Task

Ask the AI Tutor, or a friend, to run through this same three-question format with you using different content: one definition question, one short snippet to read, and one scenario. Practice answering all three in one sitting, the way a real interview would flow.

Quick Recap

• A realistic interview mixes definition, code-reading, and scenario questions in one conversation, not just one type at a time.
• The same "state it, then name the specific detail" structure works across all three question types.
• Being able to both identify a problem in code and produce the corrected version is stronger than identification alone.

Try Asking the AI Tutor

Use the AI Tutor to run a full mock interview. For example, try asking:

• "Can you run a short mock IBM i interview with me, mixing a definition question, a code snippet, and a scenario question?"
• "Can you give me a snippet with a missing %found check and ask me to both spot it and fix it?"