Self-recording
Self-recording
Record video and audio in-browser. Watch yourself back - the most uncomfortable, most useful step in interview prep.
Deepmock has two modes: quick solo reps when you have 10 minutes, or a full MMI circuit when you want the real thing. Both record. Both give feedback. Both build the muscle.
Start with a single question or run a full circuit. The feedback loop is the same either way.
Every session opens with a performance summary - what you did well, what to focus on next, and an overall score. Below it, every note pins back to the second it happened in the transcript.
Core criteria are set by the station category each has its own marksheet. On top of that, every station adds rows tuned to what it specifically tests. Each row gives you the score, and what to practise next.
A 45-year-old presents to your GP clinic with intermittent abdominal pain over the past six weeks. They describe it as dull and unpredictable, with no clear trigger. Vitals are within normal range and they have no significant past medical history. Walk me through how you would approach this patient, from the moment they sit down to your initial management plan.
Frame the problem before solving it - structured history first, then targeted exam, then investigations, then plan. The strongest answers also map the patient's ICE (ideas, concerns, expectations) and psychosocial context, not just the biology. Skipping this reads as exam-mode pattern matching, which examiners notice almost immediately.
Each step should follow from the last with no contradictions. A common trap is calling for a 'targeted exam' immediately after admitting the symptoms are vague - if you don't know what you're targeting, the exam can't be targeted. Narrate the logic out loud so the examiner can follow why you moved from one step to the next.
Patient-centered openings - introducing yourself, checking what they prefer to be called, signposting what you're about to do - build trust before any clinical content lands. Filler words like 'uh' and 'um' fragment the rhythm of your reasoning and read as uncertainty even when your content is solid. Slow down rather than fill silence.
Name specific differentials grouped by system rather than gesturing vaguely at 'sinister causes'. For abdominal pain that might be cardiac (ACS referring to the epigastrium), GI (peptic ulcer, biliary, appendicitis), MSK, or gynae. Examiners read breadth as a signal of how deeply you've thought, not just how widely.
Screen for red flags early and revisit them as your differential narrows - this is the difference between safe and dangerous practice. Strong candidates also build in a safety net: telling the patient explicitly what would warrant return and what timeline to expect. Don't treat red flags as a one-time checklist at the start.
Send a recording to other applicants or your own tutor with one link. They drop comments straight on the session - no account, no extra tools, no losing feedback in a chat app.
Every add-on and feature we build is designed to close one gap between where you are and where you need to be on interview day.
Record video and audio in-browser. Watch yourself back - the most uncomfortable, most useful step in interview prep.
Teamwork, communication, ethics, empathy, motivation, critical thinking, healthcare issues, and de-technicalise - written by practising doctors so the scenarios feel like the real thing, not textbook exercises.
See your score per domain over time. Spot the weak ones early, drill them, watch the curve climb.
Share your recording with other applicants or tutors and get structured feedback - all in one place. Free on every plan.
Get scored against the actual MMI rubric - structure, reasoning, communication, and content. See exactly where your answer held up and where it fell apart. Skip it if you'd rather work with peers.
See one way to approach the question - a full example response plus key points broken down by category. Useful for understanding what a strong answer covers, not just how it sounds.
Simulates a real Zoom-style interview - a live persona across the table, full timer, no second takes. Multiple interviewer personas available. Beta - request access via Discord.
See the emotions, micro-habits, and visual cues your face communicates throughout your answer - sentiment timeline, emotion breakdown, and habit detection like frowning or repetitive cues.
Analyses the emotional texture of your speech - confidence levels, emotional tone, filler patterns, and speaking rhythm. Get specific recommendations to sharpen how you come across.
No card, no signup wall - just a real prompt, a real timer, and a recording you can keep.
Start a Free Station