As of 13 August 2026, AI can practise coding interview questions.

What to actually do

1. Hand it to a person

   The route this page recommends

   A person who owns the outcome does this end to end, worth it when the failure is dear.

2. Do it yourself

   Second choice

   A chat interface, chat-fluent skill, and roughly 30 minutes until you can act on the result.

   How to actually do it

   1. Choose the role level, programming language and topic you need to practise, then open a chatbot such as ChatGPT, Claude or Gemini.
   2. Paste the prompt and replace each bracketed slot with your role, level, language, experience and topic.
   3. Read the problem aloud, ask the clarifying questions you would ask in an interview, and state your proposed approach before writing code.
   4. Paste your code into the chat in a code block and ask the model to test it against the stated examples and additional edge cases without giving the full solution.
   5. Run the code yourself in a local editor or coding platform, compare the results with the model's claimed output, and check the time and space complexity independently.
   6. Ask for the final assessment, record the specific mistake or improvement, and repeat with a new problem under a timed interview-style limit.

   Prompt

   Act as a realistic coding interview interviewer. I am preparing for a [job title or level] interview using [programming language]. My current experience is [beginner, intermediate or advanced], and I want to practise [topic, such as arrays, graphs, SQL or dynamic programming].
   
   Give me one interview-style problem at a time. Do not show the solution or give an unsolicited hint. Start by stating the problem, constraints, expected input and output, and two or three examples. Wait for my clarifying questions or approach.
   
   When I submit an approach or code, respond as an interviewer first: ask relevant follow-up questions about correctness, edge cases, complexity and trade-offs. Then assess it using this structure: correctness, failing cases, time complexity, space complexity, code quality, communication and one specific improvement. If the code is wrong, identify the smallest useful clue rather than rewriting it immediately. Give a full solution only when I ask for it or after I have made a serious attempt. Keep the difficulty appropriate for [target company or interview format, if known]. Use only facts that follow from the problem, and say when a judgement is subjective.

   Open it prefilled in ChatGPT or Claude, or copy it into Gemini, which takes no prefill link.

3. Use a tool built for this

   The distant third