AI Voice Interview Screening for softgarden: Pre-Screen Candidates as a Connected Add-On

You use softgarden as your ATS. It works. Your team knows it. Yet you’re searching for a softgarden video interview option because the first screening step still eats time: CV scans, phone screens, scheduling, notes, and the constant “who is worth a first conversation?” question.

This page describes a connected add-on from an external provider: Sprad + Atlas. It is not a native softgarden feature and it does not replace softgarden. Instead, Atlas docks onto your existing flow and adds a short, automated voice pre-screen upstream. Candidates answer a 3–4 minute voice interview (no CV upload required). Atlas transcribes, scores against your criteria, flags AI/bot spam, and syncs a transparent shortlist back into softgarden. If you want to see what that “Voice Apply” step looks like, start with Atlas Apply.

Why softgarden users look for a softgarden video interview step

Softgarden is a leading DACH ATS used by over 2,000 companies. On its integrations page, softgarden highlights a built-in “2-minute video introduction” feature for candidate pre-selection and integrations with video providers like Cammio, plus options such as Microsoft Teams for live interviews (softgarden integrations).

That covers two common needs:

Asynchronous introductions (short candidate videos) to get a first impression.
Synchronous interviews (Teams, etc.) to run structured conversations with hiring managers.

So why do teams still search for a softgarden video interview add-on?

Because the pain usually isn’t “we can’t do interviews.” The pain is volume and prioritisation. You can collect a short video. You can run Teams interviews. But your recruiters still need to:

review every application manually to decide who deserves time,
screen out low-effort or irrelevant submissions,
prevent bot-driven or AI-mass applications from clogging the pipeline,
keep criteria consistent across recruiters and locations,
push decision-ready summaries to hiring managers fast.

A classic “softgarden video interview” setup often starts too late in the funnel. You only invite a subset. The real workload happens before that invitation.

What Sprad + Atlas adds on top of softgarden (without replacing it)

Sprad is an AI-first HR platform used by employers including Zalando, Dior, LVM, Bijou Brigitte, and public-sector organisations such as the City of Stuttgart. For this use case, the key component is Atlas, an AI coworker that runs workflows across your people stack via a People Data Knowledge Graph.

For recruiting teams on softgarden, Atlas can be configured as an integration layer that:

adds a short voice pre-screen as candidates apply,
scores answers against your job requirements using a transparent rubric,
detects suspicious submissions (text-to-speech, behavioural patterns, honeypots),
writes results back into softgarden so softgarden stays the system of record.

Think of it like this: softgarden stays your pipeline. Atlas becomes your fast “first conversation,” completed for every applicant within minutes.

How the softgarden video interview add-on works (step by step)

The workflow is simple on purpose. You want fewer clicks, not another dashboard. The exact technical hook depends on your softgarden setup and what interfaces you use (career site embed, application form flow, ATS triggers/notifications, or agreed integration endpoints). The operational pattern stays consistent:

Step 1: The candidate hits the pre-screen at the right moment

You choose where the short interview happens:

On your career site before the candidate enters the ATS pipeline, or
Right after application submission, as an immediate next step.

This is where many teams originally searched for a softgarden video interview. The difference: instead of asking for a time-consuming video recording, Atlas can run a fast voice Q&A that most candidates finish in one sitting.

Step 2: Atlas runs a structured voice interview (about 3–4 minutes)

Candidates answer a small set of role-specific questions. You control the questions and the scoring criteria. Typical question types include:

availability / shift patterns / location (frontline roles),
must-have skills and examples (knowledge work),
language level and customer situations (service roles),
motivation and role expectations (to reduce early churn).

Because it’s voice-first, candidates don’t need perfect formatting, a long cover letter, or a CV upload for this step.

Step 3: Atlas transcribes, scores, and explains the result

Atlas converts the voice answers into text and evaluates them against your requirements. The output is designed for decision-making, not novelty:

Transcript you can read quickly.
Score per criterion (for example: availability match, required skill evidence, language confidence).
Short explanation tied to the candidate’s actual answers.
Flags for suspicious patterns (see anti-spam below).

This matters for trust. Hiring managers don’t want a black-box “yes/no.” They want to see what the candidate said and why the system scored it that way.

Step 4: Atlas pushes the shortlist back into softgarden

Softgarden remains your source of truth. Atlas writes the outputs back into the candidate record so your recruiters can work where they already work. Depending on your preferred workflow, that can include:

adding the transcript and score summary to the profile,
tagging candidates (for example “high match,” “review,” “low match”),
sorting or filtering candidates by structured scores,
optionally triggering next steps when a threshold is met.

If you also want downstream actions (scheduling, candidate emails, reminders), Atlas can run those as workflows across tools. That’s the broader “automation layer” concept behind Sprad Automate: you define the rule once, and it runs across your stack.

Softgarden video interview vs. a voice pre-screen: which problem are you solving?

When teams search for “softgarden video interview,” they often mean one of three things:

“We need richer signals than a CV.”
“We need to screen faster.”
“We need to stop low-quality / automated applications.”

A video interview can help with the first point. It can also increase effort for candidates and reviewers. A short voice pre-screen is usually strongest for the second and third point: it’s quick to complete, quick to review, and easier to standardise at scale.

Voice screening also changes the funnel shape. You can run it for every applicant, not just the few you invite after manual CV screening.

That said, voice is not “magic.” It introduces its own governance questions (bias, accessibility, documentation). The practical way to run it responsibly is to keep the scoring focused on job-related content, keep humans in control of decisions, and log the rationale and data used.

Anti-AI spam shield: stop bot applications before they reach your softgarden pipeline

Many teams don’t only want a softgarden video interview. They want protection from low-effort mass applications that waste recruiter time and distort funnel metrics.

Atlas Apply includes an anti-spam layer designed for the new reality of automated submissions. In this workflow, Atlas can flag patterns such as:

Text-to-speech (TTS) detection: signals that an “answer” was played back rather than spoken naturally.
Behavioural fingerprinting: repeated patterns across many submissions.
Honeypots: traps that normal candidates never trigger, but bots often do.

These flags shouldn’t auto-reject people. They should help you prioritise review. That’s the goal: let your team spend attention where it counts, while keeping decisions explainable.

Before/after: softgarden alone vs. softgarden + Atlas voice screening

The point isn’t to criticise softgarden. Softgarden already covers core ATS needs and offers video-related options and integrations. The point is to remove the manual bottleneck that sits around those features.

Hiring step	Softgarden-only workflow	Softgarden + Atlas voice pre-screen (connected add-on)
Initial screening	Recruiters read CVs/notes and decide who gets a first call.	Every applicant completes a 3–4 minute voice Q&A; Atlas scores against your rubric.
Candidate introduction	Optional short intro video (or manual phone screen) still needs human review time.	Transcript + scored summary is readable in seconds; you can still invite top candidates to video/live interviews.
Spam / low-effort filtering	Mostly manual detection; recruiters notice patterns late.	Anti-AI spam flags (TTS, behavioural fingerprints, honeypots) surface risk early for faster triage.
Consistency	Different recruiters interpret CVs differently; criteria drift happens fast at volume.	Same questions, same scoring logic, same documented rationale per candidate.
ATS documentation	Notes often stay unstructured; hiring managers get mixed detail levels.	Structured fields (scores, tags, summaries) are pushed back into softgarden for clean filtering and handoffs.

Two realistic use cases for teams searching “softgarden video interview”

The best use cases are high-volume roles and roles where written applications look identical. That’s where a short, structured conversation adds signal fast.

Use case 1: Frontline hiring with high applicant volume

You hire for roles with clear must-haves: availability, location, shift work comfort, basic language, right-to-work checks, maybe a certificate. CVs often don’t answer these reliably. Phone screens become repetitive.

With Atlas voice screening connected to softgarden, you can ask those must-have questions once and get consistent answers for every applicant. Recruiters don’t need 3–5 minutes per candidate just to confirm basics.

Sprad positions this workflow as saving roughly 54 minutes per candidate compared to manual screening plus a first interview step (Sprad benchmark from Atlas Apply materials). Even if your internal baseline is lower, the direction is predictable: the time moves from “talking to everyone” to “talking to the shortlist.”

What changes operationally?

You review a ranked list inside softgarden, not a raw inbox of applications.
Hiring managers get a consistent summary and can pick interview slots faster.
Candidate communication becomes easier because decisions happen sooner.

Use case 2: Knowledge-work roles flooded with polished applications

In many office roles, written applications have converged. Candidates use templates. Some use AI heavily. A “softgarden video interview” step might help, but it still costs time to review and invites bias toward presentation quality.

A short voice pre-screen gives you something harder to fake at scale: a real-time answer to a specific prompt tied to your role. Atlas can flag suspicious voice patterns and still preserve fairness by keeping the scoring criteria job-related and visible.

The best pattern here is to ask for evidence:

“Tell us about a project where you owned X. What was the outcome?”
“What’s your approach when stakeholder Y pushes back?”
“Which tools have you used weekly in the last 6 months?”

Then you only invite the top segment into a deeper softgarden video interview or live interview loop.

Why an integration layer beats adding yet another recruiting tool

Most teams don’t fail at hiring because they lack software. They fail because work gets duplicated across tools: ATS, calendars, email templates, spreadsheets, chat, and hiring manager follow-ups.

The Sprad/Atlas design choice is to sit as an automation and intelligence layer on top of what you already use. Sprad describes it as “one AI for your entire HR stack,” driven by a People Data Knowledge Graph. Practically, that means Atlas can read status from tools and write results back, instead of forcing your team into a new system for each workflow.

If you want to understand that integration concept beyond recruiting, the most direct overview is Atlas integrations across 1,500+ tools. The recruiting use case is one workflow. The strategic value is that you can reuse the same layer for scheduling, onboarding orchestration, manager nudges, and other routines later.

Avoid the “rip-and-replace” trap

Replacing softgarden is a big decision. Data migration, process redesign, training, works council questions, and the risk of losing hard-earned adoption can slow you down for months.

A connected module approach keeps the ATS stable and improves the part that hurts most: the first screening bottleneck.

Keep your operational rhythm

Recruiting teams build muscle memory around one system. Softgarden pipelines, tags, stages, and hiring manager access are part of that rhythm. An add-on only works if it respects it.

That’s why the “write back into softgarden” part matters more than any AI feature. If recruiters need to jump between tools, you’ll lose the time you hoped to save.

Commercial model: setup project, then usage-based AI costs

Most recruiting add-ons are priced like SaaS: per seat, per recruiter, per vacancy, per month. That can punish you for scaling hiring capacity.

Sprad’s Atlas automation model is positioned differently: typically a one-time setup project (often 2–4 weeks, depending on scope and interfaces), then ongoing AI/API usage costs. No per-seat SaaS license is the core idea for Atlas-driven automation workflows.

What does the setup usually cover?

Defining the screening questions and scoring rubric per role family.
Aligning on what is “must-have” vs. “nice-to-have” vs. “red flag.”
Integration design: where the pre-screen sits and how results sync to softgarden.
Pilot calibration with a limited set of jobs to validate outcomes.
Governance: retention, access permissions, and review processes.

This approach fits best when you want measurable workload reduction without a long platform replacement program.

DACH lens: Datenschutz (GDPR/DSGVO) and works council readiness

If you operate in DACH, the first questions are predictable: What data do you collect? Where is it stored? Who can access it? How do we document decisions? Does the works council need to be involved?

This page can’t provide legal advice, but it can outline practical guardrails teams commonly use when they add an automated screening step:

1) Data minimisation and purpose limitation

Collect only what you need for the screening purpose. A short voice Q&A can be configured to focus on job-related criteria rather than personal topics. That helps with internal acceptance and reduces risk.

2) Transparency for candidates and internal stakeholders

Make the process understandable: what the pre-screen is, how long it takes, what it influences, and that final decisions remain with humans. Atlas is designed to provide transcripts and scoring rationale so reviewers can trace outputs back to actual answers.

3) Access controls and auditability

In regulated environments, you want clear role-based access: who can listen to audio, who can see transcripts, and how long data is retained. An audit trail for changes and decisions helps you answer “why did we invite/reject?” months later.

4) Works council involvement (when applicable)

Many organisations involve the Betriebsrat early when introducing new tooling that affects hiring workflows. The earlier you align on purpose, safeguards, and human oversight, the smoother the rollout tends to be.

If you want to see how Atlas is positioned as an HR-native layer across tools (not just recruiting), the broader workspace context is explained on Sprad’s Atlas workspace.

FAQ: softgarden video interview add-on (voice screening edition)

Is this a native softgarden video interview feature?

No. Atlas Apply is a third-party module. It connects to your softgarden workflow and writes results back so you can stay in softgarden for daily recruiting work.

Do candidates need to upload a CV for the voice pre-screen?

No. The voice step is designed to work without a CV upload. You can still collect CVs later in the process if your roles require it.

Does voice screening replace a softgarden video interview?

It can replace the first screening call for many roles. Many teams still use a softgarden video interview or live interview for the shortlist. The win is that you stop spending human time on the bottom of the funnel.

How do we avoid a black-box “AI says no” process?

You keep humans in control and require transparency: transcript visibility, criterion-level scoring, and documented reasons. You can also set the process so Atlas doesn’t auto-reject candidates, but prioritises review.

How does the anti-spam shield help with AI-generated applications?

It flags suspicious patterns (for example text-to-speech playback and bot-like behaviours) so your recruiters can prioritise genuine submissions. The goal is triage, not automated exclusion.

Where Atlas can go next (once the softgarden screening bottleneck is fixed)

Teams usually start with screening because that’s where workload spikes. Once an integration layer exists, you can automate more of the busywork that slows time-to-hire, still keeping softgarden as your ATS.

Common next steps include:

interview scheduling coordination across calendars and hiring managers,
structured interview scorecards and consistent documentation,
personalised rejection emails at scale (with approvals),
onboarding orchestration triggered when a candidate becomes a hire.

Those workflows sit under Sprad Automate, where the promise is simple: you design the workflow once, then it runs across your tools.

A practical way to evaluate a softgarden video interview alternative without disrupting your ATS

If your goal behind “softgarden video interview” is faster screening, cleaner shortlists, and fewer spam applications, evaluate the workflow with three checks:

Time-to-decision: How quickly can your team produce a shortlist after applications arrive?
Signal quality: Do you get job-related evidence you can explain to hiring managers?
ATS cleanliness: Does everything land back in softgarden without manual copy-paste?

That’s the core promise of an integration-first approach: keep what already works (softgarden), and remove the manual screening bottleneck with a connected module. For the specific voice screening piece, the most direct reference is Sprad Atlas Apply. For the wider “integration layer across your HR stack” concept, review Sprad integrations.

Jürgen Ulbrich

CEO & Co-Founder of Sprad

Jürgen Ulbrich has more than a decade of experience in developing and leading high-performing teams and companies. As an expert in employee referral programs as well as feedback and performance processes, Jürgen has helped over 100 organizations optimize their talent acquisition and development strategies.