Python coding test practice

What the real test usually looks like

There is no single Python coding test. Each employer chooses a testing platform and then sets its own questions, number of tasks, time limit and allowed languages. Your test invitation is the best guide to yours. Public examples show the range:

• HackerRank. HackerRank lists its Python (Basic) skills test as 2 questions in 1 hour 30 minutes, covering topics like scalar types (HackerRank). Its help center lists the question types an employer can put in a test, including coding, multiple choice, database, code review and subjective questions, and says employers can add questions from HackerRank's library or write their own (HackerRank Help).
• TestGorilla. TestGorilla's Python coding tests page says most of its individual coding tests take 15 to 35 minutes, lists an entry-level algorithms test at 15 minutes, and says candidates can solve problems in Python or in other languages the employer allows (TestGorilla).

Parts commonly reported in employer Python tests:

• Coding tasks: write a function in an online editor, run it against sample inputs, then submit it to be checked against more test cases you may not see.
• Multiple choice: Python concepts, what a snippet prints, and which code is correct.
• Debugging or code review: find and fix the problem in code someone else wrote.
• Sometimes more: a SQL question, a larger project task, a take-home assignment, or webcam and screen monitoring during the test.

What our mocks cover. Our full mocks have 20 questions in 60 minutes in four sections: Python concepts (6), what the code prints (6), finding the bug (5) and building a solution (3). All code is Python 3. You can go back, change an answer and flag questions until you submit. Our mocks practice reading, tracing, debugging and choosing code. They do not run code you write. Many real assessments do ask you to write code that runs against hidden test cases; this practice builds the reading, tracing and debugging skills behind that, so add some timed practice in a real editor as well. Our mocks also leave out proctoring, plagiarism checks, video interviews, take-home and project tasks, and any employer pass mark.

Question types with worked examples

One worked example for each section, with the trap most people fall into.

1. Python concepts. What is the difference between a == b and a is b? The first asks whether two values are equal; the second asks whether they are the same object. Two separate lists [1, 2] are equal but not the same object. Trap: using is to compare numbers or strings. It can seem to work for small values and then fail.

2. What the code prints. x = [0] * 3 then x[1] = 5 and print(x). Trace it: [0, 0, 0] becomes [0, 5, 0]. Now try grid = [[0] * 2] * 2 then grid[0][0] = 1. It prints [[1, 0], [1, 0]], because both rows are the same inner list. Trap: reading the code as you expect it to behave, not as Python runs it. Write down each variable's value line by line.

3. Finding the bug. A loop for i in range(len(items) - 1): is meant to print every item but skips the last one. The bug is the - 1: range(len(items)) already stops before the length. Trap: fixing the line that crashes or looks odd rather than the line that causes the wrong result. Test the code in your head with a tiny input, such as a list of two items.

4. Building a solution. You need the count of each word in a list. collections.Counter(words) does it in one step; a dictionary with counts[w] = counts.get(w, 0) + 1 in a loop does it by hand. Trap: picking the option that handles the normal case but not the edges. Check every choice against an empty list, one item, repeats and negative numbers.

How to prepare

1. Read your invitation first. Find out which platform you will use, how many tasks there are, the time limit, whether you can choose Python 3, and whether the test is monitored. Formats vary too much to assume.
2. Try the platform's practice area if it has one. Get used to the editor, the Run button and how sample test cases are shown, so the screen is not new on the day. HackerRank, for example, describes candidates clicking Run to run the test cases while they work (HackerRank Support).
3. Trace code by hand. For short snippets, write each variable's value after every line. This is the core skill behind output and debugging questions, and it is how you check your own code without running it.
4. Know the everyday tools. Lists, dictionaries, sets, strings, slicing, sorted versus list.sort, enumerate, zip, list and dictionary comprehensions, and collections.Counter.
5. Think about edge cases before you submit. Empty input, one item, repeated values, negative numbers, very large inputs. Hidden test cases often check exactly these.
6. Write some code against the clock. Our mocks train reading and debugging. Also solve a few small problems in a real editor with a timer, so writing and running code under pressure feels normal.
7. Practice at the real pace. If your test is like HackerRank's Python (Basic) example, 2 questions in 1 hour 30 minutes gives about 45 minutes a problem (HackerRank). Take full timed mocks so pacing yourself feels normal.

How scoring usually works

Scoring depends on the platform and the employer. There is no universal scale and no universal pass mark. Coding questions are usually scored automatically against test cases. Two public descriptions:

• HackerRank says coding questions are evaluated with predefined test cases that measure correctness and efficiency. Each test case has its own score: code that passes a test case receives that test case's full score, and code that fails it receives zero. The question's total is the sum of the test cases passed, so a solution that passes some test cases earns part of the points (HackerRank Support). Coding and multiple choice questions are evaluated automatically, while some types, such as subjective answers and file uploads, are evaluated by hand (HackerRank Help).
• TestGorilla describes percentile-based scoring based on overall correctness, meaning whether given inputs produce the correct output, and a report showing results for each test case (TestGorilla).

What counts as a good result is up to each employer, and many look at your code as well as the score.

Our practice score is the number of our own questions you answered correctly, with a breakdown by section, so you can see where to focus. It is a practice score: it is not a score on any platform's test and does not predict whether you will pass.

Common mistakes

• Mixing up methods that change a list and functions that return a new one. list.sort() and list.append() return None; sorted() returns a new list.
• Forgetting that assignment shares objects. b = a does not copy a list or a dictionary. Use a[:], list(a) or copy.deepcopy(a) when you need a copy.
• Off-by-one errors in ranges and slices. range(n) stops before n, and s[1:3] takes two items, not three.
• Starting a search with a made-up value. Starting a maximum at 0 or a minimum at 1000 breaks on unusual inputs. Start from the first item, or from float("-inf").
• Only testing the example. The visible example usually passes. Hidden test cases check empty input, single items, duplicates and large inputs.
• Running out of time on one task. If you are stuck, submit a simple version that passes some test cases, move on, and come back if time allows.