The defense is a conversation. Rehearse for the conversation.
Defense Preparation Coach turns the final weeks before your PhD defense from passive re-reading into a structured rehearsal program. You build a question bank across eight research-grounded categories, draft answers with live speaking-time estimation so they land in the 1:30–3:00 target, and rehearse aloud through a self-rated practice loop that resurfaces your weakest answers more often.
Your questions, answers, and notes never leave your browser. No AI reads them; nothing is uploaded; no microphone is touched.
You've mastered the content. You haven't practiced the defending.
The dissertation defense is a speaking event. The cognitive load of constructing a clean, time-bounded, confident answer under committee pressure is a separate skill from knowing your material — and almost nobody practices it. Re-reading the thesis builds recall; it does not build the muscle the defense actually tests.
- Answers either ramble or get cut short. Without a target window, you can't tell whether your answer is "right" — only whether it landed in 90 seconds or eight minutes. Both fail.
- The question bank lives only in your head. So you rehearse the same five questions you remember, and freeze on the categories you didn't.
- Vulnerabilities go unnamed. If you haven't said "the sample size in Study 2 is small, here's how I defend it" aloud, at speed, you will not say it well under pressure.
A coach, not a workbench
How it works
Set your defense date
A countdown anchors the workspace; a "today's suggested action" surfaces the smallest next step — add a question, draft an answer, run a practice rep.
Build the bank, draft the answers
Capture questions across the eight categories. The drafting editor estimates speaking time live at 130 wpm and flags when an answer drifts outside the 1:30–3:00 window.
Rehearse aloud, rate yourself
Practice Mode hides your draft, runs the timer, then asks you to score yourself on three honest dimensions. A weighted queue resurfaces weak and stale answers more often.
What's inside
Eight-category question bank
Capture, edit, sort, filter, and search your own questions across the eight categories committees draw from. Tag each to a category, a difficulty, and (optionally) an expected examiner.
Live speaking-time estimation
The drafting editor reads your answer at the academic speaking rate (130 wpm) and updates the speaking-time estimate as you type, with a non-blocking compress warning when you drift past 3:00.
Practice mode with self-rating
Full-side-panel takeover: surface a question, hide the answer, time you, then prompt three sliders — covered the key points, stayed on time, sounded confident. Sessions accumulate.
Weighted practice queue
Lower-rated and longer-untouched questions weigh more heavily in the queue, so the answers you most need to rehearse come up most often — without ever fully retiring a mastered one.
Vulnerability map & examiner profiles
Unlimited entries for the weaknesses you fear most (with a prepared response) and for each committee member's role, expertise, and likely angles. Tag questions to the examiner most likely to ask.
No AI, on purpose
Nothing here generates your questions, drafts your answers, or evaluates your reasoning. There is no microphone access, no recording, no transcription. The substantive work is yours; the structure and timing are the tool's.
Privacy & Trust
- Your defense prep never leaves your machine. Questions, draft answers, vulnerability notes, examiner profiles, and practice ratings all live in IndexedDB and
chrome.storage.localon your device. - No AI runtime, no telemetry, no analytics, no account. The extension does not read any web page and runs no inference on your content.
- No microphone, no recording, no transcription — ever. Practice mode is a stopwatch and a self-rating dialog, not a voice analyzer. The brand rule is permanent.
- One optional network call, only if you ask for it. If you enter an email and tick "Get occasional defense-prep tips," only that email address (plus a
defense-preparation-coachtag) is sent to GradSummit's sign-up service — never any question text, answer text, vulnerability note, or examiner detail. Unsubscribe link in every email.
It is a coach for time-bounded defense rehearsal — not an AI examiner, and not a guarantee of outcome. Read the full privacy policy.
Built by Dr. Rafiq Muhammad, PhD — author of the Mastering Research book series.
Pricing
Every feature, for everyone — no subscription, no tiers, no account.
The eight-category question bank, the live speaking-time estimator, practice mode with self-rating, the weighted queue, and unlimited vulnerabilities and examiners are all unlocked.
Email is entirely optional and opt-in — leave it blank forever, or add it to hear about occasional updates.
Frequently asked questions
How is this different from Defense Prep Workbench?
The Workbench ships ~150 curated anticipated questions across 10 categories, an 8–12 week structured arc, and three deterministic exports (defense prep brief, Q&A flashcards, morning summary) — it is the structural workbench. The Coach is the rehearsal companion: you write your own questions across the eight categories committees draw from, draft tight answers with a live speaking-time estimate, and rehearse aloud against a clock with self-rating. They pair well — the Workbench gives you the scaffold; the Coach drills the delivery.
Does any AI generate my answers or predict my examiner's questions?
No. There is no AI runtime anywhere in this tool. The eight categories are static, hand-written taxonomy; the questions are yours; the answers are yours. The speaking-time estimator is a plain word count divided by a 130 words-per-minute constant — no model, no inference. The premise is that the structure plus honest reps are what produce a defended defense.
Why no microphone or recording?
Three reasons. First, brand rule: we never request microphone access in any GradSummit extension, ever. Second, audio is not what improves defense delivery — honest self-rating is. Third, the moment a tool starts recording you, you over-prepare for the recording, not for the room. The stopwatch + slider loop is the minimum that drives the calibration you actually need.
I have less than 2 weeks until my defense. Can this still help?
Yes — the tool is calendar-agnostic. Set the date, the countdown anchors the workspace, and the suggested action surfaces the highest-leverage next step (draft an answer for one of your gaps, then run a practice rep on it). Two weeks of disciplined daily reps beats two months of re-reading.
Is my defense prep content uploaded anywhere?
No. Your questions, answers, vulnerability notes, examiner profiles, and practice ratings all stay in IndexedDB and chrome.storage.local on your device. The only thing the extension can send to GradSummit is your email address — and only if you explicitly opt in to product updates and tick the box. No defense content is ever transmitted. See the full privacy policy.
How much does it cost?
It's free, with no subscription and no paid tiers. Every feature is available to everyone. There's nothing to buy and no account to create.
What does the "weighted practice queue" actually do?
Every question has two signals: how recently you practiced it and the rolling average of your self-rating. The queue weights questions with lower ratings and longer-ago last-practice more heavily, so the answers you most need surface most often — without ever fully retiring a mastered question, since under pressure even strong answers slip.
What if I find a bug?
Email support@gradsummit.com — replies go to a real human (the developer).