Apex Interview Questions and Answers

1What it is

Governor limits are hard caps on what one transaction may use: how many queries it runs, how many records it reads and saves, how much CPU time and memory it takes. Cross one and Salesforce stops the code with a LimitException.

2Why it matters

Salesforce is multi-tenant: thousands of companies share the same servers. The limits make sure one org's badly written code cannot slow everyone else down. They also force code to be written for real data volumes - a trigger has to cope with 200 records at once.

3Example

4In short - say this in the interview

1What it is

Bulkified code handles many records at once instead of one at a time: it collects what it needs into sets and maps, runs one query and one DML for the whole batch, and never puts SOQL or DML inside a loop.

2Why it matters

A trigger does not get one record - it gets up to 200 at a time from a data load, an integration or a mass update. Code that queries once per record works in a demo and fails at the 101st record in production.

3Example

for (Contact c : Trigger.new) {
    Account a = [SELECT Name FROM Account WHERE Id = :c.AccountId];
}

Set<Id> ids = new Set<Id>();
for (Contact c : Trigger.new) ids.add(c.AccountId);
Map<Id, Account> accounts = new Map<Id, Account>(
    [SELECT Name FROM Account WHERE Id IN :ids]);

4In short - say this in the interview

1What it is

A before trigger runs before the record is written to the database; an after trigger runs once it has been written, but before the transaction commits.

2Why it matters

The timing decides what you can do. Before the save you can change the record's own fields for free - no DML. After the save the record has its Id and system fields, so it is the place to create or update related records.

3Example

4In short - say this in the interview

1What it is

A before insert trigger on Contact that rejects any new contact whose email already exists - bulkified, so a load of 200 contacts runs one query.

2Why it matters

Duplicates break reporting and email sends. Rejecting them before they are saved costs no extra DML, and addError shows the user why, on the field itself.

3Example

trigger ContactDuplicateCheck on Contact (before insert) {
    Set<String> emails = new Set<String>();
    for (Contact c : Trigger.new) {
        if (c.Email != null) emails.add(c.Email.toLowerCase());
    }

    // One query for the whole batch, outside the loop.
    Set<String> existing = new Set<String>();
    for (Contact c : [SELECT Email FROM Contact WHERE Email IN :emails]) {
        existing.add(c.Email.toLowerCase());
    }

    for (Contact c : Trigger.new) {
        if (c.Email != null && existing.contains(c.Email.toLowerCase())) {
            c.Email.addError('A contact with this email already exists.');
        }
    }
}

4In short - say this in the interview

1What it is

The three collections: a List is ordered and allows duplicates, a Set is unordered and holds each value once, a Map stores values by key.

2Why it matters

Choosing the right one is most of bulkification. Sets collect Ids for a query without duplicates; maps let you find a record by Id instantly instead of looping through a list.

3Example

List<Contact> toUpdate = new List<Contact>();          // order kept, DML on the whole list
Set<Id> accountIds = new Set<Id>();                     // each Id once, ready for WHERE Id IN
Map<Id, Account> byId = new Map<Id, Account>(
    [SELECT Id, Name FROM Account WHERE Id IN :accountIds]);
Account a = byId.get(someContact.AccountId);           // instant lookup, no loop

4In short - say this in the interview

1What it is

Keywords on an Apex class that decide whether it respects the running user's record-level sharing. with sharing enforces it, without sharing ignores it, and inherited sharing takes on the mode of the code that called it.

2Why it matters

Apex runs in system mode by default - it can see records the user cannot. The keyword is how you stop a user interface from showing records the user has no right to, while still letting a system process see everything when it must.

3Example

4In short - say this in the interview

1What it is

Three ways to run Apex later, in the background: a @future method, a Queueable job (System.enqueueJob) and Batch Apex (Database.executeBatch).

2Why it matters

Background work gets higher limits and does not keep the user waiting - and some work, like a callout after a save, can only happen there. Each option trades simplicity for control and volume.

3Example

4In short - say this in the interview

1What it is

Deploying Apex to production needs at least 75% overall code coverage, every test passing, and every trigger covered. Test.startTest and Test.stopTest mark the part of a test that is the code under test.

2Why it matters

Coverage is Salesforce's guarantee that code is exercised before it reaches production. startTest gives the code a fresh set of governor limits, so test setup does not use them up; stopTest runs any queued async work immediately, so you can assert on its result.

3Example

@isTest
private class ContactDuplicateCheckTest {
    @isTest
    static void blocksASecondContactWithTheSameEmail() {
        insert new Contact(LastName = 'First', Email = 'a@example.com');

        Test.startTest();
        Database.SaveResult result = Database.insert(
            new Contact(LastName = 'Second', Email = 'A@example.com'), false);
        Test.stopTest();

        System.assert(!result.isSuccess(), 'The duplicate should be rejected');
    }
}

4In short - say this in the interview