Screen every applicant, read only the best.

The match score is built from the same things you would check by hand: whether the candidate's skills, experience, and background line up with what the role actually requires, with synonym and typo tolerance so wording differences do not cost someone unfairly. Every score comes with the specific reasons behind it, so you are never asked to trust a number blind, you can see exactly why someone ranked where they did and decide for yourself if you agree. Think of it as a fast, consistent first pass across every applicant, not a replacement for your judgment on the candidates who make the shortlist. You still make every hiring decision. Aster just makes sure that decision is not being made on an incomplete read of the pile.

No. Auto-screening does the first pass, the part where you would otherwise open 50 CVs just to find the 10 worth a closer look. Once the shortlist is ranked with reasons attached, you still review the actual candidates, their full profile, and any scorecard the rest of the team has filled in, before making a call. What changes is where your hour goes: instead of spending it opening every application in application order, you spend it on the ones already shown to be a plausible fit. The final read, the interview, the offer decision, all of that is still yours to make, Aster just clears the pile down to what is actually worth your attention first.

Yes, that is exactly what bulk-upload is for. You can upload a batch of CVs you already have, in whatever format they are already sitting in, and Aster parses and scores each one the same way it would a fresh application. They become part of your searchable talent pool, queryable by skill or experience, with the original source tracked so you know where each candidate came from. Instead of that backlog sitting unused because searching it by hand was never worth the time, it becomes something you check before you even post a new role, which sometimes means filling a position without waiting on new applicants at all.

You send a shortlisted candidate a scheduling link. They pick a slot from the availability you have set, and Aster books it, creating a Google Meet or Microsoft Teams link automatically and syncing with your Google or Microsoft calendar so there is no double-booking. You do not propose times, wait for a reply, or manually send an invite. Reminders about the upcoming interview go out automatically by email and WhatsApp ahead of time, so neither side shows up unprepared or forgets. The only manual step left in the whole process is showing up to the interview itself and actually talking to the candidate.

For a lot of candidates, especially anyone who does not check a work or personal email closely, WhatsApp gets read faster than email does. That is the reason it is built in as a reminder channel alongside email, not instead of it. Templated reminders go out automatically ahead of interviews and at other points in the pipeline, so candidates get a nudge somewhere they are actually likely to see it before it is too late to reply. The template wording is editable, so it matches how your team actually talks to candidates, it is not locked to a stiff, generic script that reads like it came from a machine.

Everything is encrypted in transit and at rest, and scoped to your workspace, nobody outside your team can see it. It is never used to train shared models, so a resume you upload does not end up shaping how the system scores someone else's company's candidates. You can export or delete candidate data at any time, which matters if a candidate asks you to remove their information or you are cleaning out an old backlog you no longer need. Role-based access and an audit trail mean you also know who on your own team looked at what candidate, and when they did it, which matters for a shared pipeline.