Zoom
Listens from the desktop
Desktop interview assistant · now in early access
The interview co‑pilot
with good instincts.
A standalone desktop app that listens to the interview on any video tool — Zoom, Google Meet, Teams, Webex — and gives you real-time probes and rubric scoring against your team's actual values. Nothing to install inside your meeting software.
Jedah keeps track
M
Maya Chen
CuesCandidate
recording
12:04
Consent on file
End
LiveSaved
Next probe
“You said the team scaled 8 to 40. What broke first, and what did you stop doing to fix it?”
Third “we” in a row on this story.
Next probe
“Walk me through the first version you shipped. What did you cut to get it out?”
Answer stayed at strategy level for 90 seconds.
Not yet covered
“When did you last change your mind because someone on the team pushed back?”
Nothing on this yet, either way.
NotesTranscript
Autosaves
Your notes…
Stage
Build rapport
Say: “Please walk me through…”
Behavior-based questions
Ask: “What do you want to do?”
Listen & record
Connect-ualizing questions
Sell & salutation
Pause
Fits where you already work
Works on whichever video tool your company runs on.
Google Meet
Nothing to install in it
Greenhouse
On Business and up
Lever
On Business and up
Microsoft Teams
Same desktop app
Webex
Same desktop app
SSO and SAML
On Enterprise
Custom rubric builder
On Business and up
How it works
Three moments, and Jedah is useful in all of them.
BEFORE
You have a job description
Paste it. The rubric and the questions come from the actual role.
Rubric drafted
Seven dimensions, weighted for a Staff Engineer on payments.
Questions generated14
DURING
You are in a conversation
Cues arrive in the pauses. You stay in the room, which is the job.
Here is something worth asking about
“Mostly” is carrying weight. Ask what shipped late.
Stage 3 of 714:04
AFTER
You need a write-up
Your notes, reconciled against what was really said. You edit it; both versions stay.
Brief ready
Evidence-linked, gaps marked as gaps rather than scored.
Generatedone pass
Ask Jedah
Ask about the candidate. Get the receipt with the answer.
Two days later, when someone asks why you rated them that way, you can ask the record instead of trusting your memory. Answers come back quoting the line they came from — and “the record does not say” when there is nothing to quote.
M
Full review
Maya Chen
Staff Engineer, Payments · Onsite
Where did she actually show ownership, and where am I filling it in?
Twice, both in the second interview. She took the payments migration after the original lead left, and she names what she got wrong — the dual-write window was too long and she owned the rollback.
The third example in your notes, the hiring freeze, is her describing a team decision. She says “we” throughout and never places herself in it.
“I kept dual writes on for six weeks longer than
I should have. That was my call and I rolled it back.”
Interview 2 · transcript
Is she a flight risk?
The record does not say. Nothing in either interview or the reference call touches tenure, why she left, or what she is looking for next.
What would answer it: a reference from her current manager, or asking her directly what would make her stay somewhere five years.
Ask a question…→
It answers from one candidate’s record
Their interviews, the transcripts, your notes and any reference calls — and nothing else. Not what is typical for the role, not what similar candidates did.
Every claim carries its quote
Answers cite the line they came from and where it appeared, so you can check the reasoning rather than take it on trust.
Silence is an answer
When the record cannot support an answer it says so and names what would settle the question — a hedged invention is still an invention.
Untested is not a weakness
A dimension nobody assessed is never reported as a low score. Not looking and looking-and-finding-nothing are different facts and stay different.
Pricing
Simple pricing. No surprises.
Professional
$499 /month, up to 5 seats
- Live desktop interview copilot
- AI rubric scoring (HPF)
- Debrief agenda generation
- NYC LL144 + Illinois BIPA compliance
- 50 interviews a month
Business
$1,999 /month, up to 20 seats
- Everything in Professional
- Team performance analytics
- Greenhouse and Lever integration
- Custom rubric builder
- Multi-panel calibration
- 200 interviews a month
Enterprise
$4,999 /month, unlimited seats
- Everything in Business
- White-label branding
- SSO and SAML
- EU GDPR + AI Act compliance
- Dedicated Slack channel and SLA
- Unlimited interviews
Flat pricing per company, not per seat. New workspaces start on a 30-day trial, no card.
Straight answers
Questions people actually ask.
Is this just a notetaker?
No. A notetaker tells you what happened once it has finished happening. A follow-up question is useful for about fifteen seconds, and that is the window this is built for. The write-up afterwards is reconciled against your own notes rather than being a generic recap.
Does it replace our ATS?
No. Greenhouse and Lever integration is there on Business and up. Your ATS runs the pipeline; Jedah works the conversation.
What if it is wrong about someone?
Then the panel should catch it. Scores are visible across the panel and calibrated before the debrief, so a single wrong read gets challenged by a second one rather than becoming the decision.
Where does our data go?
Into your workspace. Biometric data is destroyed per policy, subprocessors are disclosed, and there is no training on your candidate data.
How long does setup take?
It is a desktop app, so you install it once and it works on whichever video tool the call happens to be on. A new workspace starts on a 30-day trial and needs no card.
Run your next interview with Jedah.
Jedah is in early access, with hand-picked design partners reshaping how their teams interview. Tell us the role you are hiring for.
Request a demo