Hear the question, see the answer in about a second written from your résumé, thesis or sales deck, not the open internet. Built for job interviews, vivas, sales calls and client meetings.
Pick from models by five providers, each labelled with its speed, whether it reads your screen and what an answer costs.
Product names and logos are trademarks of their respective owners. Use does not imply affiliation or endorsement.
inneedai is a real-time AI interview and meeting assistant: a Windows desktop app that picks up the other side of the conversation from your computer's audio — no bot joins the call — transcribes it live and, on a hotkey, puts an answer on screen — or on your phone — in about a second. It runs job interviews, viva examinations, sales calls and client meetings, each grounded in the material you upload for it — a résumé, a thesis, a deck or your own notes.
Why did you choose variational inference rather than standard MCMC for estimating posterior parameter variance?
In Chapter 3 (pp. 38–41), variational inference enabled linear scaling O(N) across our 2.4M-token dataset where MCMC suffered from autocorrelation time τ > 450. Empirical Bayes checks in Table 4 confirm posterior accuracy within 1.2%.
[1,2]Thesis_Final.pdf· 2 passages| Who it is for | Job interviews, viva examinations, sales calls and client meetings. |
|---|---|
| What it hears | Your computer's audio output — the other side of the call. Your microphone is never opened. |
| How fast | An answer on screen in about a second, from a hotkey. |
| Where you read it | In a small overlay on your screen — or on your phone, with Second Screen. How Second Screen works |
| What it runs on | Windows 10 and 11, 64-bit. A macOS build is in development. |
| What others see | Nothing joins the call. With stealth mode on — it ships off — the overlay is left out of screen shares. What that covers |
It hears the question, finds the answer in your own material, and puts it where only you can see it — while the conversation is still going.
Upload a résumé, a thesis, a deck or your notes. Answers are argued from them, and say which document — and where — they came from.
The other side is transcribed as they speak, so the answer is ready the moment they stop.
Marks the overlay as protected content, so a screen share leaves it out. It ships switched off — you turn it on.
It listens to your computer's audio output. Nothing is added to the meeting, and your microphone is never opened.
It answers the questions it recognises on its own. For anything else, your hotkey brings an answer up.
Second Screen puts the question and the answer on your phone as they stream. Scan a QR code — no app to install — then tap Answer now or Clear on the phone while the Windows app listens.
How Second Screen works
It works out which language is being spoken and keeps up when a call switches mid-sentence. There is no setting to change.
Pick the session type and both the tone and the material it draws on change with it.
The question you hoped they would skip, answered in your own words and your own history — not a generic script.
Your résumé and the role you pasted in
AI interview copilotA thesis defence runs on recall under pressure. Answers are argued from your own paper, and say which part they came from.
Your thesis — answers cite where they came from
Viva preparationObjections and pricing questions arrive without warning. Answers come from your own material, so nothing gets promised that you cannot deliver.
Your deck and pricing — never an invented number
AI copilot for sales callsA status call turns into a hard question about a detail three documents deep. The answer is already on screen.
The brief you uploaded, as live talking points
Assistant for client meetingsCan you walk me through a difficult technical trade-off you owned and how you explained it to leadership?
When decoupling our monolith's billing engine, I chose an event-driven ledger over dual-write locks. I presented a trade-off matrix to leadership showing that eventual consistency avoided distributed deadlocks, achieving 99.995% transaction reliability.
Setup takes about two minutes. Most of it is choosing which file to upload.
A résumé, a thesis, a deck, a brief. It is parsed for you — no forms to fill in.
The role, the topic, the agenda or the deal, so answers fit this conversation.
The desktop app transcribes the call in the background. No bot, nothing to see.
Grounded in your own material and on screen in about a second.
A chatbot can't hear the call, and a note-taker answers after it ends. inneedai is built for the moment in between.
| What you need | inneedai | A chatbot in another tab | A meeting note-taker bot |
|---|---|---|---|
| Hears the question from the call itself | YesFrom your computer's audio | NoYou type or paste it | YesRecords the meeting |
| Answers while the call is still going | YesAbout a second | PartlyAs fast as you can type | PartlyBuilt for notes and summaries |
| Argues from your own résumé, thesis or deck | YesUploaded per session, with sources | PartlyOnly what you paste in | NoNot what it is for |
| Stays out of the attendee list | YesNo bot joins | YesSeparate window | NoJoins as a participant |
| Can be kept out of a screen share | YesWith stealth mode on | NoShows if you share it | NoIt sits in the call itself |
Monthly plans you can cancel any time — or buy credits and pay only for what you use.
Try inneedai on live calls, free — top up credits for more time.
15 mins/month · Fast Groq models
No credit card required
For an active job search.
6 hrs transcription · 1,000 AI credits
Then from your AI credit balance · cancel any time
For back-to-back rounds and live deals.
40 hrs transcription · 5,000 AI credits
Then from your AI credit balance · cancel any time
What the other side sees, what it hears, and what it costs.
No. inneedai runs four kinds of session — job interviews, viva examinations, sales calls and client meetings — and each one is grounded in different material. An interview answers from your résumé and the role you pasted in; a viva answers from your own thesis and says which part it drew on; a sales call answers from your deck and pricing; a client meeting answers from the brief you uploaded. The framing changes with the session type, so a viva answer argues and a sales answer sells.
Your computer's audio output — the other side of the call. Your microphone is never opened, so what you say is not transcribed and not sent anywhere.
Nothing joins the call, so there is no bot in the attendee list and no participant to notice. What they can see is your screen, if you share it. An optional stealth mode marks the overlay as protected content at the operating-system level so a screen share does not contain it — but it ships switched off, and turning it on is your decision.
What the other side seesThe desktop app runs on Windows 10 and 11, 64-bit. A macOS build is in development and the download button for it is disabled until it ships.
You choose, from models by five providers. Free includes three fast models on Groq; holding credit adds OpenAI and DeepSeek models; Pro adds Google Gemini; Power adds Anthropic's Claude and the largest OpenAI model. Before a session, each model shows how quickly it starts answering, whether it can read your screen, whether it thinks first and roughly what an answer costs. Or leave it on your plan's default and we pick for you.
Eighteen, and it works out which one is being spoken on its own — there is no language setting. It also keeps up when a call switches between two of them mid-sentence, which is the case that usually breaks live transcription.
The desktop app listens to your computer’s audio and puts a grounded answer on your screen in about a second — and with stealth mode on, it stays out of anything you share.
Windows 10 & 11 · 64-bit macOS coming soon