The AI agent that
drives your interviews.
RecruitSense joins your Google Meet and runs the interview with you — navigating your question playbook, pushing sharper follow-ups, and flagging suspicious answers in real time. Even non-technical recruiters can confidently run technical screens.
Chrome Web Store listing in review · Invite-only access during beta · Works inside Google Meet
Maya Chen · Senior Backend Eng
You · Recruiter
RecruitSense agent
Backend · Round 2 · 2/5 answered
Question playbook
Auto-navigating- Walk me through your last system you owned end-to-end.· Ownership
- Tell me about a production outage you debugged.· Operational
- Design a rate limiter for our public API.· System design
- How would you decide between SQL and a key-value store?· Tradeoffs
- Describe a time you mentored a junior engineer.· Leadership
Suggested follow-up
How would you handle bursty traffic from a single tenant without blocking everyone else?
- Runs inside Google Meet
- Real-time transcription
- Whole panel, one live copilot
- Structured report in 30 seconds
Built so any recruiter can run any interview — even the technical ones.
How it works
Hand the wheel to an agent that actually knows the job.
You write the playbook once. The agent runs it with you on every call.
01
Build the playbook
Paste the job spec. RecruitSense drafts a structured question list with the exact answers a strong candidate should hit. Edit it in two minutes — or pick from a template.
02
Install, join the Meet
We ship the Chrome extension directly to design partners while the Web Store listing is in review. The agent panel appears the next time you open a Google Meet.
03
Drive the interview together
The agent navigates your playbook in real time, suggests sharper follow-ups against the candidate's last answer, and flags possible cheating.
04
Get a structured report
Hit End Interview. The recording assistant leaves instantly, and within 30 seconds every question gets a pass/fail with quotes, a match score, and a recommendation.
What the agent does
Three jobs, one fluid interview.
Click through what the agent is doing on the other side of the panel during a 45-minute Google Meet.
Question playbook · 2 / 5
- Walk me through a system you owned end-to-end.
- Tell me about a production outage you debugged.
- Design a rate limiter for our public API.
- How would you decide between SQL and a key-value store?
- Describe a time you mentored a junior engineer.
On now
Design a rate limiter for our public API.
Expected signals — checked off live as the candidate hits them.
- Clarifies scope (per-user vs. per-IP vs. per-endpoint).
- Picks an algorithm and justifies it (token bucket, leaky bucket, sliding window).
- Names a backing store and discusses TTL / atomicity.
- Handles burst traffic and noisy-neighbor scenarios.
- Plans for the store going down (fail-open vs. fail-closed).
The recruiter unlock
Run a technical screen even when you don’t know the right answer.
The agent already knows what a strong answer looks like for this role — because you wrote the playbook. You listen, watch the verdict, decide.
Candidate · 0:14:23
“I’d shard with rendezvous hashing instead of consistent hashing — better load balance under churn, and the math is actually simpler.”
Maya Chen · Senior Backend Engineer candidate
Agent · checking the rubric
- Picks an appropriate sharding algorithm.
- Justifies the tradeoff (load balance under churn).
- Discusses key distribution math.
- Mentions collision handling.
What you see
✓ Strong on system design
2 of 4 sub-signals covered. Push next on: “How would you handle hash collisions?”
You don’t need to know
- How rendezvous hashing differs from consistent hashing.
- Whether Redis is the right backing store for this case.
- If their CAP-theorem explanation is correct.
The agent does. So you stay focused on the candidate — not on keeping up.
Built for panels
Your whole panel. One live copilot.
Interviews aren’t solo work. Panelists join the same live AI chat — asking, verifying, and steering together, with every message attributed to the person who sent it.
Panelists on every interview
Add teammates to an interview when you schedule it. Everyone assigned gets the live copilot chat — lead interviewer and panel see the same playbook, the same flags, the same AI.
One shared chat, fully attributed
Every question, message, and AI answer is tagged with the teammate who sent it — live in the call, and again in the final report. No more "who asked that?"
Invite your team by email
Send branded invites with Admin, Recruiter, or Viewer roles. Teammates sign in with Google and land in your workspace — no setup call required.
The bot cleans up after itself
When you hit End Interview, the recording assistant leaves the Meet immediately — no robot lingering in an empty room — and the full recording stays available for playback.
Live copilot chat
Backend · Round 2 · 3 interviewers in the room
Daniel (Eng Manager)
Her sharding answer sounded rehearsed — can we verify it?
Room access is enforced server-side — only the lead interviewer and assigned panelists can join.
What the agent does
An agent built to drive the call, not just record it.
RecruitSense isn’t a notetaker. It’s a co-interviewer that knows the role, runs the playbook, and pushes back when something looks off.
Structured Question Navigator
Define the questions you want to ask once — with the answers a strong candidate should give. The agent walks you through them in the call, tracks which ones are covered, and lets you mark them answered with one click.
Proactive follow-ups
The agent reads every answer against your rubric and surfaces a sharper next question — citing exactly what the candidate just said. Junior interviewers ask like seniors.
Cheating detection
Catches the tell-tale signs: gaze drifting off-screen, long suspicious pauses, answers lifted verbatim from a top search result. The agent flags the moment and tells you what to re-ask.
Recruiters can run technical loops
Because the agent already knows what a strong answer to a system-design question looks like, a non-technical recruiter can run a technical screen — and score it the same way a senior engineer would.
Ask the agent anything, live
Type a question into the panel mid-call — “is her Redis answer actually right?”, “what should I push on next?” — and get an answer in seconds, grounded in the live transcript.
Per-question scorecard
Hit Leave and within 30 seconds you get a pass / partial / fail on every playbook question, with the candidate's exact words as evidence and a recommended next step.
Why not a notetaker?
Notetakers help after the call.
RecruitSense helps during it.
A transcript tells you what happened. An agent in the call changes what happens.
From the founder
I started RecruitSense after watching too many interviews go off the rails — recruiters reading questions they didn't fully understand, candidates clearly looking up answers, hiring managers writing reports from memory days later.
The tools out there were notetakers. They recorded the call and handed you a transcript. None of them helped during the conversation — when help would have actually mattered, and when an AI agent that knows the role can change the outcome.
So we built a co-interviewer that drives the call with you: it runs your playbook, pushes the right follow-up against the candidate's last answer, flags when something looks suspicious, and writes the structured report before you stand up. The result: any recruiter on the team can run any interview.
Sanzhar Danybayev
Founder, RecruitSense
Pricing
Free while we’re in beta.
We’re onboarding our first cohort of design partners. You get the whole product — we get your feedback.
Beta access
$0/ forever during beta
- Unlimited Google Meet interviews
- Structured Question Navigator
- Proactive AI follow-ups (Claude)
- Cheating detection + flags
- Per-question scorecard reports
- Panelists + shared live copilot chat
- Team invites with roles (Admin / Recruiter / Viewer)
- Candidate kanban across your team
Chrome Web Store listing in review — we onboard design partners directly. We’ll introduce paid plans before we charge anyone, and early users get grandfathered.
FAQ
Questions hiring teams ask us.
Yes. A recording assistant joins the Google Meet visibly, the way familiar notetaker bots do — candidates see it as a participant. When you end the interview, it leaves the call immediately. We believe in transparent interviews; the agent's guidance panel is only visible to your side.
Put an AI co-interviewer in your next Google Meet.
Join the private beta — your next interview gets a structured playbook, live follow-ups, cheating flags, and a scorecard you didn’t have to write.
Free during beta · No credit card · Chrome Web Store listing in review