Frequently Asked Questions

EngMock is an AI technical interview practice platform covering System Design, Coding/DSA, OOD, and AI-assisted coding. A live voice interviewer runs each interview in the workspace its format needs: an architecture whiteboard for system design, a code editor for algorithms and a multi-file workspace for object-oriented design and AI-assisted coding. After the interview you get a scored report with the evidence behind each score.

Every coding question (algorithm, object-oriented design and AI-assisted coding) can be answered in any of nine languages: Python, JavaScript, TypeScript, Java, C++, C#, Go, Rust and Kotlin. You pick the language in the editor once the interview starts, and you can switch at any time; the interviewer follows along.

Yes, in Runnable Coding: Run executes your entry point and any test files you write in an isolated sandbox with no network access, and the interviewer sees every run. Algorithm and object-oriented design questions also offer Whiteboard Coding (No Run), with no autocomplete and no Run, where you validate with written tests and a verbal walkthrough. AI-assisted coding interviews always use Runnable Coding.

Yes. Live OOD scenarios use voice plus a multi-file workspace in any of the same nine languages (TypeScript by default). The interviewer probes requirements, responsibilities, APIs, extensibility, validation and design trade-offs, and may change a requirement mid-interview.

You work in a multi-file repository with EngMock’s own coding assistant. It reads the repository but not the interview question, the rubric or the interviewer, and it cannot edit or run your code. It is the only AI that counts: the interview judges how you frame requests, review and adapt what it suggests, verify the result and explain the final code, from your code, your runs, your conversation with EngMock’s assistant and what you explain to the interviewer. Two of the 6 questions start from an existing codebase: you fix its defects, build a new feature and then make it fast. Assistant output is never counted as your own work.

Unlike a generic chatbot, EngMock runs a timed, track-specific interview, follows your whiteboard or code as you work, asks adaptive follow-ups, and produces evidence-based feedback with a rubric designed for that interview format.

A computer with a microphone and a modern web browser (Chrome recommended). The whiteboard and the code editor run in the browser, so you do not need to install anything. VS Code mode is optional: it needs VS Code 1.90 or later and the EngMock extension, which you download from EngMock.

EngMock currently offers 100 live mock interviews: 43 system design, 25 algorithm, 26 object-oriented design and 6 AI-assisted coding. They are separate from the Question Bank, which publishes paraphrased candidate-reported questions without solutions.

You choose 45 or 60 minutes when you set up the interview. You can end it early at any time, and the report is generated after the interview ends.

Yes, and we encourage it. Each session is saved with its transcript, your whiteboard or code, and your scores, so you can compare attempts on the Sessions page.

It reads the structure of what you draw rather than a picture: each labelled box, text note and arrow, and which component every arrow connects. That description reaches the interviewer when you pause, so its follow-ups can point at a specific component or connection, such as a single database on a hot path.

Yes. In My Scenarios you can write a scenario yourself or let AI draft one from a question or your notes, edit it, and then practice it live with the same interviewer and reports. Custom scenarios can be system design, algorithm, object-oriented design or AI-assisted coding interviews. Creating one needs purchased interview credits or an active Unlimited plan; Free Credits cannot be used for custom scenarios.

Yes, in VS Code mode (preview). For algorithm, object-oriented design and AI-assisted coding questions you choose VS Code on the setup page, install the EngMock extension from the .vsix file EngMock provides, and connect with a one-time pairing code. You write and run code in VS Code while the voice interview stays in your browser tab. VS Code mode always uses Runnable Coding. Inline completion from your editor (for example GitHub Copilot suggestions) is allowed; AI chat assistants and agents are not. In AI-assisted coding interviews, EngMock’s coding assistant opens in its own column to the right of your code, and it is the only AI that counts: your AI use is assessed from your conversation with it.

Yes. EngMock has an MCP server at https://api.engmock.com/api/mcp. Connect a client such as Claude, Claude Code, Codex, GitHub Copilot CLI or GitHub Copilot in VS Code to create and edit your scenarios, read your interview history and reports, and get a mock interview suggested with its setup link, a checklist and its cost. Each token or connected app gets only the permissions you choose. Interviews themselves always run on the website, and this is separate from the AI-assisted coding track, where you code with EngMock’s assistant during the interview.

Each track has its own rubric. System design covers requirements, estimation, APIs and data, architecture, deep dives, reliability, trade-offs and communication. Algorithm interviews emphasize problem solving, correctness, complexity, implementation, testing and communication. OOD focuses on modeling, responsibilities, APIs, extensibility, implementation and validation. AI-assisted coding also evaluates how you delegate, review, correct and verify the assistant’s work and whether you own the result.

You choose a Junior, Mid, Senior or Staff bar. Reports say whether you met it, estimate the level you demonstrated, and include specific strengths, evidence-backed improvement areas and actionable next steps.

An overall score out of 100, a score out of 10 for each skill with the reasoning behind it, the result against your target level, phase coverage, quotes from your answers, what a strong answer looks like, and a replay of the session along its timeline. You can save it as a PDF or download it as Markdown.

You can pay for live AI mock interviews in two ways. Interview credit packs are one-time purchases that never renew: one credit covers one complete 45 or 60-minute interview, and each pack also includes timed access to the Candidate-Reported Question Bank (Starter 30 days, Value 60 days and Ultimate 90 days). The optional Unlimited plan is a subscription instead: US$9.99 a month, renewing until you cancel, for unlimited interviews, one at a time, on your own Gemini API key. Google bills that key for model usage, usually about US$1.50–4 per full 45–60 minute interview.

No. Question Bank access is included with every paid credit pack (Starter 30 days, Value 60 days and Ultimate 90 days) and with the Unlimited plan while it is active. There is nothing extra to buy.

Yes. New accounts get two Free Credits: each covers one complete live interview on a Free Credit eligible scenario (Find a Pair Matching a Target or Design a URL Shortener), with its full report. The Unlimited plan itself has no free trial.

Yes. Purchased credits are valid for a fixed time after purchase: Starter and Value credits for 90 days and Ultimate credits for 180 days. The Pricing page shows the current packs.

EngMock works in current versions of Chrome, Firefox, Safari and Edge. We recommend Google Chrome for the best audio and whiteboard experience.

Yes. All interview sessions are encrypted in transit and at rest. We do not sell your data. We share it only with the providers that run the service, such as Google Gemini for the AI interviewer and feedback, with tools you connect through MCP, within the permissions you approve, or when the law requires it. Other users cannot see your interviews, transcripts or custom scenarios. See our Privacy Policy for full details.

EngMock is designed for desktop and laptop use because of the whiteboard and the code editor. We recommend a device with a larger screen.