The first screen, done properly, for every applicant.
Agni calls your applicants, works through the questions that matter for the role, and records availability and expectations as clean fields. Your recruiters start the day with a shortlist and a transcript for every name on it, not a list of numbers to dial.
This is what a slow screen costs you.
Illustrative scenario voices, not customer testimonials.
The first screen is a volume problem.
Most first calls ask the same ten questions: are you still interested, can you do the shift, when can you start, what are you expecting. It is careful work that does not need a recruiter's judgement, and it arrives in bursts nobody can staff for. Seasonal peaks, new sites, campus intakes, a contract win that needs sixty people by next month. Agni takes the repetition so your recruiters keep the judgement.
*Language and dialect coverage figures are forward-looking, pending verification in the claims audit, and scope-dependent. Calling windows are configured per deployment and per market.
Built for the first screen, not just the first hello.
It takes over the dialling and the repetition. The shortlist decision stays with your recruiters.
New applicants in your ATS or careers form start a screening call through the API or a webhook. For a hiring drive, upload a list and run it as a campaign with calling windows, retry rules and a cap on concurrent calls.
One question set per role or job family. Agni confirms interest, then asks about experience, skills, licences or qualifications the role needs, and follows up when an answer is vague. Every candidate for the same role gets the same questions.
The answers that sink interviews late are captured on the first call, as structured fields, not scribbled notes from a rushed phone screen.
Every screen ends with a summary, a disposition and the full transcript sent to your ATS by webhook. Strong candidates go straight to a recruiter by live transfer or a booked callback. Everyone else hears a clear next step and a polite close.
From applicant list to recruiter-ready shortlist.
One agent handles the first screen from trigger to handoff, and every call leaves a structured record behind it.
Knockout rules, pass criteria and handoff triggers are yours to define. Agni ranks and routes; a person makes every shortlist decision.
Candidates answer better in the language they think in.
A nervous applicant in their second language gives short, safe answers. The same applicant in their first language tells you what they actually did in their last job. Agni switches with them mid-sentence and keeps the screen on track.
Illustrative conversation. *Simulated progress meter.
The decision stays human.
Agni gathers the answers and hands them over. Your recruiters review, decide, and own every shortlist.
Transcripts, recordings, summaries and dispositions arrive as a payload your team maps to the ATS fields you already report on. Custom API work is in scope on the enterprise plan.
Recordings and transcripts on every screen, your retention policy, role-based access for recruiters and hiring managers, and a complete audit trail.
All product and company names are the trademarks of their respective owners. *SLA figures are indicative pending the claims audit and are agreed per deployment.
Priced by the minute, not by the seat.
A first-round screen is a short call. You pay for the minutes Agni spends talking to candidates, so a quiet month costs less than a hiring drive. Bring your volumes to the demo and we will work through a realistic year one with you.
*Indicative pricing, not a proposal or offer, plus a platform subscription from $150 a month, depending on concurrency. Setup may apply depending on complexity, and pricing varies by region and solution. Enterprise custom builds are negotiated.
Design, deploy, operate. With your recruiters in the loop.
Your roles, your screening questions, your knockout rules and your disclosure wording. We agree the languages, the calling windows in each market, and the fields each screen must return. Your hiring lead signs off the script before a single candidate is called.
Webhooks wired into your ATS, test calls on the team's own phones, then a shadow period on a slice of real applicants while your recruiters read every transcript and check every disposition against their own judgement.
Full volume, a named engineer, and a regular review of completion rate, disposition accuracy, handoff quality and candidate drop-off. New roles and new languages ship as changes, not new projects.
Deployment timelines depend on integration complexity, telephony access, the number of roles in scope, and how quickly script sign-off happens on your side.
Every candidate gets the same fair first call.
Hiring is a sensitive place to put AI, and candidates deserve to know how they are being screened. The controls below are part of every deployment, not add-ons. Your legal and people teams set the policy for each market you hire in; we build the calls to match it.
These are product controls, not legal advice. Hiring rules differ by country, state and province; confirm your requirements with your own counsel.
The questions your talent lead will actually ask.
Yes. Each role or job family gets its own questions, knockout rules, voice and call behaviour. We build the first version from the questions your recruiters already ask, and you can test it with calls to your own team before a real candidate hears it. Changing a question later is an edit, not a rebuild.
By webhook. When a call ends, Agni sends the transcript, recording link, summary and a structured disposition, and your team maps that payload to your ATS fields during deployment. Calls can also be started through our API, so a new applicant in your system can trigger a screen automatically. We do not claim a native connector for any particular ATS; webhooks are how it connects.
Yes, and we recommend it in every market. The call opens with a disclosure that the caller is an AI assistant calling on your behalf. Consent is asked and logged, calls stay inside the windows you set, and every screen keeps its recording and audit trail. A candidate who would rather speak to a person is offered one.
They get one. Agni can transfer the call live to a recruiter with the screen so far attached, or book a callback and record the request as a structured outcome. Nobody is told to email a general inbox and wait.
The model handles 50+ languages and 200+ regional dialects,* and it follows a candidate who switches languages mid-answer. Most hiring teams start with the two or three languages their applicants actually speak and add more when a new site or market opens.
Yes. Upload a list and run it as a campaign with schedules, calling windows, retry rules and a limit on concurrent calls. It suits seasonal peaks, new site openings, graduate intakes, and contract wins that need a lot of people quickly.
In the region you choose. Data residency is part of the deployment design, you set retention on recordings and transcripts, and access is role-based so hiring managers see only their roles. We hold SOC 2 (in process), HIPAA and GDPR controls, and your security team gets a working session with our engineer, not a PDF.
Pricing starts from $0.06 per minute* plus a platform subscription from $150 a month, depending on concurrency, with setup depending on complexity. Enterprise custom builds are negotiated and include custom API work. Bring your application volumes to the demo and we will work out a realistic year one together.
You hire a handful of senior people a year. Executive and specialist search runs on relationships and judgement from the first conversation. An AI first call adds nothing there, and we will say so in the demo rather than after you sign.
You want AI to make the hiring decision. We will not build that. Agni gathers answers, ranks and routes. A person reviews every shortlist, and that is not a setting you can switch off with us.
Your applicants all speak one language and your recruiters already call back within a day. The dialect advantage is the reason to choose us. If it does not apply, a simpler single-language tool is a fair choice.
Your ATS cannot receive a webhook and nobody will export the data. Structured outcomes are only useful if they land somewhere. Sort out where the records will go first, then talk to us.
*All figures on this page are forward-looking, pending verification in the claims audit, and indicative only. Pricing is pending final confirmation.
One live call. The speaker flips language mid-sentence and Agni keeps going without missing a beat.
HEAR IT LIVE → Try itTalk to Agni yourself. Ask it a screening question in your own language and listen to how it answers.
HEAR IT LIVE → Book itBring your talent lead and one real role. We run a screen on it, live, in the language your applicants speak.
BOOK A DEMO →Ready to hand off the first screen?
Bring your talent lead and one open role. We screen a test candidate in your applicants' language, live on the demo.