An AI consultation simulator is only as good as the case set you practise, and the default behaviour — replaying the dramatic scenarios you enjoy — builds fluency on a skewed sample the real exam does not share. This workflow, the implementation companion to our Geeky Medics UKMLA simulator audit, shows how to build a balanced UKMLA case set across the content map, preserve unseen cases for calibration, and keep the clinical knowledge inside each consultation current.
What you are working with
Geeky Medics' AI virtual patients let you consult by voice or text across history-taking, information-giving and counselling scenarios, with AI automarking and post-scenario vivas; its scenarios are mapped to the MLA conditions and presentations. Verify the current case count and price on the product page, and note the stated limit: it supports communication-skills stations, not physical examination or procedures. The relevant exam component is the CPSA — the clinical and professional skills assessment no MCQ bank touches.
Why a balanced case set matters
Consultation skill is partly general and partly presentation-specific: the structure transfers, but the content, red flags and management differ by presentation. Practise ten cardiology histories and you will consult cardiology fluently and freeze on a mental-health presentation you never rehearsed. The exam samples across the content map, so your case set should too — which means deliberately building a rota, not following the platform's suggestions or your own preferences.
Building the rota
Map your practice across the content map's breadth. Ensure every area of clinical practice appears — including the ones you avoid, which are usually mental health, sexual health, safeguarding-adjacent and the "difficult conversation" scenarios that feel uncomfortable to rehearse. Balance common against rare, acute against chronic, and pure history-taking against counselling and information-giving. A simple discipline: keep a checklist of the map's areas and tick each as you practise a case in it, forcing yourself into the under-practised corners rather than the comfortable ones.
Preserve unseen cases
Because the case library is finite, ring-fence a reserve of never-attempted cases from the start. Replaying cases rehearses those specific consultations rather than building transferable skill, so your fluency becomes case-specific. Save the reserve for the final fortnight, when a cold case is the only honest test of whether your structure, timing and questioning transfer to a presentation you have not memorised.
Keep the knowledge current
The simulator rehearses the consultation; it does not guarantee the medicine inside it is current. A candidate who consults fluently but gathers toward an outdated management plan loses marks the communication cannot recover. So pair each block of cases with a knowledge check on the management and red flags the cases surfaced — this is where iatroX sits in the stack, as the current-knowledge layer, not as a consultation simulator.
A worked example
Suppose your first fortnight's cases cluster on chest pain, breathlessness and abdominal pain — the classic history-taking staples. The rota discipline flags the gap immediately: no mental-health presentation, no sexual-health history, no safeguarding-adjacent scenario, no difficult-conversation case. Next fortnight forces those, one per session, with the vivas reviewed for the knowledge each exposes. By preserving a handful of cases across all these areas for the final week, you keep a cold test of whether your structure holds outside your comfort zone.
A seven-day pattern for UKMLA candidates
Monday: two history-taking cases from under-practised map areas, timed to station length, feedback reviewed against observable behaviours. Tuesday: one counselling or difficult-conversation case plus a peer sanity-check of the AI's scoring. Wednesday: a timed, unseen 40-question mixed block in iatroX's free UKMLA bank to measure the knowledge underpinning your consultations. Thursday: one case in a new area; review the viva and fill knowledge gaps. Friday: two cases rotating the library. Saturday: a mixed session — one simulator case plus a timed MCQ block — reviewed by error type. Sunday: rest, preserving unseen cases. Geeky Medics rehearses the consultation; iatroX keeps the knowledge inside it current; neither pretends to be the other.
Continue, supplement, switch or stop
Continue while your case-set breadth and knowledge base both improve. Supplement with in-person, examiner-observed practice for the physical-examination and non-verbal skills the simulator cannot assess. Switch only if the case mix or feedback proves unreliable. Stop replaying familiar cases in the final fortnight; calibrate on preserved unseen scenarios and confirm the underlying knowledge with unseen MCQ blocks.
A worked example: spotting a skewed case set
Suppose your first fortnight's Geeky Medics practice, following your own inclinations, clusters on chest pain, breathlessness and abdominal pain — the classic history-taking staples that feel like productive practice. The rota discipline flags the gap immediately: no mental-health presentation, no sexual-health history, no safeguarding-adjacent scenario, no difficult-conversation case, and nothing from the paediatric or older-adult ends of the map. That skew is invisible while you practise, because each individual case feels useful; it only shows up when you set your case log against the content map's breadth. Next fortnight forces the missing areas, one per session, and by preserving a handful of cases across all of them for the final week, you keep a cold test of whether your structure holds outside your comfort zone.
Why a balanced set beats more reps
The temptation with a finite case library is to do more repetitions of the cases you find satisfying, and it is the wrong instinct. Consultation skill has a general component (the structure) and a presentation-specific component (the content, red flags and management), and only the general component transfers freely; the specific component has to be built presentation by presentation. Ten reps of chest-pain histories over-train a structure you already have while leaving the mental-health consultation you have never rehearsed to exam day. A balanced set of first attempts across the map builds more transferable skill than a deep pile of reps in a few comfortable areas — which is why the rota, not the rep count, is the thing to optimise, and why the unseen MCQ blocks that check the underlying knowledge matter as much as the consultations themselves.
Frequently asked questions
Is Geeky Medics enough for UKMLA on its own? No — it rehearses communication-skills stations well, but it does not cover physical examination, procedures or the applied knowledge test, so it is one component of a CPSA-plus-AKT stack.
Which UKMLA component does Geeky Medics not reproduce well? Physical examination, procedures, non-verbal communication and examiner variability in the CPSA — and it does not address the AKT at all, which needs a question bank.
How many unseen Geeky Medics cases should I preserve for final UKMLA calibration? Keep a reserve of several never-attempted cases across different map areas for the final fortnight, so your last practice measures transfer to cold scenarios rather than recall of rehearsed ones.
When should I stop using Geeky Medics and move to mixed mocks? When your case-set breadth is complete and your knowledge base is stable, shift the final stretch to full timed CPSA-style practice (ideally examiner-observed) and preserved unseen cases, using the simulator for warm-up.
How should I combine Geeky Medics with iatroX without duplicating practice? Use Geeky Medics to rehearse a balanced set of consultations and iatroX to build and measure the knowledge inside them — differentials, red flags, management and prescribing — with unseen MCQ blocks confirming that knowledge transfers.
The bottom line for UKMLA candidates
The honest one-line verdict on Geeky Medics for the UKMLA CPSA: a genuinely useful rehearsal tool for the communication half of the exam, whose value depends entirely on the case set you build and the knowledge you keep current inside it. Replay your favourites and you rehearse a skewed sample; build a balanced rota across the content map and you build transferable consultation skill. Trust the automarking as a verdict and you overcorrect on what it cannot see; use it for the observable half and get human eyes on the rest, and it earns its place. And remember what it does not do: physical examination, procedures, non-verbal rapport, and the applied knowledge test entirely. Geeky Medics rehearses a balanced set of consultations; iatroX keeps the knowledge inside them current and measures whether it transfers; human observers judge the rapport — three jobs, honestly divided.
Editorial notes and references
Written by Dr Kolawole Tytler, NHS GP and founder of iatroX. Last checked 19 July 2026; Geeky Medics figures are vendor-published — verify counts and price on the product page. UKMLA CPSA format is per the GMC and your medical school. Disclosure: iatroX operates a UKMLA question bank but is not a consultation simulator, and this workflow says so plainly. Corrections via the feedback route on iatrox.com. References: GMC MLA and CPSA guidance (gmc-uk.org); Geeky Medics product pages (geekymedics.com); related reading: the Geeky Medics UKMLA simulator audit and why your Q-bank percentage is not your exam score.