ATS Integration for AI Hiring: Darwinbox, Keka, and Zoho Recruit Guide

A practitioner's guide to connecting AI phone screening and AI interviews with Darwinbox, Keka, and Zoho Recruit — without double data entry or broken hiring workflows.

Why ATS Integration Matters for AI Hiring

Your ATS is the system of record for every hire. The moment you introduce an AI hiring tool that sits outside it, you create a gap — and that gap has a cost.

Without ATS integration for AI hiring in India, here's what typically happens: a recruiter downloads a shortlist from the AI screener, manually uploads it into Darwinbox or Keka, then copies notes across systems when the interview is complete. If the candidate drops off mid-process or the AI interview surfaces a red flag, the ATS never knows. The pipeline breaks silently.

The result is predictable: HR teams get fatigued, data gets stale, and the speed advantage of AI hiring evaporates inside the manual handoff.

With a direct integration, the flow runs end to end. A candidate applies in your ATS, gets automatically enrolled in an AI phone screening or AI interview, and the results — scores, transcripts, disposition — write back to their profile in the ATS without anyone touching a spreadsheet.

This matters especially at scale. When Indus Towers screened 10,000 applicants in 48 hours, the only way that was operationally feasible was because every screened candidate flowed directly back into their hiring pipeline. Manual reconciliation at that volume would have taken a separate team working for days.

The three ATSes that matter most for enterprise and mid-market hiring in India are Darwinbox, Keka, and Zoho Recruit. Each has a different architecture, and the integration approach for each is slightly different. Here's what actually works.

Darwinbox + AI Interview Integration

Darwinbox is the dominant HRMS for large Indian enterprises — used across BFSI, manufacturing, retail, and conglomerates. Its recruitment module (Darwinbox Recruit) supports both API-based integrations and webhook triggers, which makes it well-suited for real-time AI hiring workflows.

How It Works

When a candidate reaches a configured stage in Darwinbox Recruit (typically "Shortlisted" or a custom stage you define), a webhook fires to BabbleBots. BabbleBots initiates the AI phone screening or AI interview automatically. Once the interview is complete, the score, summary, and a link to the full transcript are written back to the candidate's profile in Darwinbox using the Darwinbox API.

Recruiters see everything in one place. No toggling between tools.

What Data Syncs

From Darwinbox to BabbleBots: Candidate name, mobile number, job ID, job title, and the hiring stage that triggered the integration
From BabbleBots to Darwinbox: AI interview score, structured summary, recommended disposition (Proceed / Hold / Reject), interview transcript link, and completion timestamp

Setup Steps (Darwinbox + BabbleBots)

Identify the trigger stage in Darwinbox Recruit — the stage at which you want AI screening to begin (e.g., "Application Received" for volume roles, "Shortlisted" for mid-funnel interviews).
Enable webhook notifications for that stage in Darwinbox under Settings → Integrations → Webhooks. Set the endpoint to the BabbleBots webhook URL provided during onboarding.
Map the candidate fields — confirm that mobile number, job ID, and candidate name are included in the webhook payload. These are required for BabbleBots to initiate the call.
Configure the BabbleBots job template — select the job from BabbleBots, assign the interview script or screening questionnaire, and set pass/fail thresholds.
Set up the write-back via Darwinbox API — BabbleBots uses Darwinbox's REST API to update the candidate record. You'll need to generate an API key in Darwinbox (Settings → API Access) and provide it to BabbleBots during setup.
Run a test candidate — push one real candidate through the configured stage and verify that the AI screening triggers, completes, and the result appears in Darwinbox within 5 minutes of interview completion.
Set stage-advancement rules — optionally configure Darwinbox to auto-advance candidates above a certain BabbleBots score to the next hiring stage.

Setup typically takes 2–4 hours with IT involvement for API key generation and webhook configuration. Darwinbox's enterprise support team can also assist if your instance is managed.

Keka + AI Interview Integration

Keka is the most widely used HRMS among Indian mid-market companies — strong in IT services, startups, and professional services firms. Its hiring module (Keka Hire) supports third-party integrations via API and has a growing partner ecosystem.

How It Works

Keka Hire uses a stage-based pipeline. BabbleBots integrates via Keka's REST API to monitor pipeline stage changes. When a candidate hits the designated stage, BabbleBots pulls the candidate record, initiates the AI interview or phone screening, and pushes results back as a note or custom field update on the candidate profile.

Because Keka's webhook support is more limited than Darwinbox's, the integration uses a polling approach by default — BabbleBots checks for new candidates in the trigger stage at regular intervals (every 5–10 minutes). For high-volume roles, this can be tightened or replaced with a Zapier/Make intermediary that provides real-time triggering.

Setup Steps (Keka + BabbleBots)

Generate a Keka API key — in Keka, go to Settings → Integrations → API Keys. Create a key with read/write access to the Hiring module.
Identify your trigger stage — this is the pipeline stage in Keka Hire that signals a candidate is ready for AI screening. Common choices: "Applied" (for volume screening) or "HR Review" (for interview-stage AI assessment).
Provide the API key and stage name to BabbleBots — during integration setup, BabbleBots maps the trigger stage to an interview template.
Configure interview templates in BabbleBots — assign the relevant job, language (Hindi or English), and evaluation criteria.
Set the write-back format — BabbleBots can write results to Keka as: (a) a candidate note, (b) a custom field if your Keka instance has been configured with custom hiring fields, or (c) a stage-change trigger if the score crosses your defined threshold.
Test with a pilot role — run 5–10 candidates through the integration before enabling for all open roles. Confirm that scores and summaries appear in the candidate profile within 15 minutes of interview completion.

Keka integration is typically lighter on IT involvement than Darwinbox. A Keka admin and a BabbleBots onboarding specialist can complete it in one session.

Zoho Recruit + AI Screening Integration

Zoho Recruit is popular among Indian recruitment agencies, IT staffing firms, and SMEs that use the Zoho ecosystem. Its strength is its native automation builder (Zoho Blueprint and Custom Functions), which gives significant flexibility for integration logic without heavy coding.

How It Works

Zoho Recruit supports both REST API and Deluge-based custom functions. BabbleBots integrates via Zoho's REST API, triggered either by a Blueprint stage transition or a custom webhook you define inside Zoho Recruit. Results write back to the candidate or job application record as a custom field or as an activity note.

Zoho Recruit's Blueprint makes this integration particularly clean: you can define the exact stage where AI screening triggers, add an automated approval gate based on the BabbleBots score, and move candidates forward or hold them — all within a single configured Blueprint.

Setup Steps (Zoho Recruit + BabbleBots)

Enable API access in Zoho Recruit — go to Setup → Developer Space → APIs. Generate a self-client OAuth token with ZohoRecruit.modules.ALL scope.
Map your hiring pipeline stages — identify which stage should trigger AI screening. In Zoho's module manager, note the exact stage name as it appears in the API (case-sensitive).
Create a Blueprint transition (recommended) — in Zoho Blueprint, add a transition into the trigger stage that calls a Zoho custom function. This function sends a webhook to BabbleBots with candidate details.
Configure the BabbleBots job in the platform — link the Zoho job/vacancy ID, assign the screening script, and set evaluation parameters.
Set up the write-back — BabbleBots calls the Zoho Recruit API to update the candidate's application record. You can configure it to write to an existing field (like "Candidate Status") or a custom field you create (e.g., "AI Screen Score").
Activate the Blueprint — once tested, activate the Blueprint so it applies to all new candidates entering that stage.
Configure score-based routing — optionally use Zoho Blueprint's conditional logic to auto-advance candidates above your threshold score and flag others for manual review.

Zoho Recruit integration is the most flexible of the three — Zoho's Blueprint allows you to build fairly sophisticated routing logic without external tools. For recruitment agencies managing multiple clients with different screening criteria, this is a significant advantage.

What to Check Before Integrating (IT + HR Checklist)

Before your IT team and ATS admin start configuration, work through this checklist. These are the questions that surface 80% of integration issues before they become problems.

Data and Access

[ ] Do you have admin-level API access to your ATS (not just recruiter access)?
[ ] Are candidate mobile numbers stored consistently in your ATS (format: +91XXXXXXXXXX)?
[ ] Is the ATS instance on a plan that includes API access? (Some Keka and Zoho tiers restrict API usage)
[ ] If your ATS is on-premise (rare for Darwinbox, uncommon for others), is the API accessible externally?

Process Alignment

[ ] Which hiring stages should trigger AI screening vs. AI interviews? (Screening = top of funnel, Interviews = post-shortlist)
[ ] Have you defined pass/fail score thresholds with the hiring manager before configuring?
[ ] What happens to candidates who don't answer the AI call — are they auto-rescheduled or held?
[ ] Does your DPDP Act 2023 compliance framework cover AI-based interview data? Candidate consent for AI screening should be captured at the application stage.

Candidate Experience

[ ] What language should the AI screening be conducted in? BabbleBots supports Hindi and English by default, with regional languages available for specific roles.
[ ] What time window should calls be placed? (Avoid early morning / late evening for non-urgent roles)
[ ] Is the candidate's mobile number the right contact channel, or do some roles route through WhatsApp?

Common Integration Issues and How to Fix Them

These are the issues that come up most frequently during ATS integration setup — and the resolution for each.

Issue: AI screening triggers but no call is placed Most common cause: the mobile number in the ATS is in the wrong format or missing the country code. BabbleBots requires numbers in +91XXXXXXXXXX format. Fix: add a field validation rule in your ATS to enforce this format at the point of application.

Issue: Results are not writing back to the ATS Check the API key permissions first. A common mistake is generating a Keka or Zoho API key with read-only access. The write-back requires write permissions on the hiring module. Also verify that the candidate record exists in the ATS at the point BabbleBots tries to write — if the candidate was deleted or merged, the write-back will fail silently.

Issue: Duplicate triggers — the same candidate gets called twice This usually happens when a stage-change event fires more than once (e.g., a recruiter moves a candidate back and forward through stages). Fix: configure a deduplication rule in BabbleBots that blocks a second interview trigger if one is already in progress or completed for the same candidate-job combination within a defined window (e.g., 7 days).

Issue: Scores are not appearing in the ATS after the interview Verify the custom field name in your ATS exactly matches what is configured in BabbleBots. Darwinbox and Zoho are case-sensitive on field names. A mismatch of even one character will cause a silent write failure.

Issue: Candidates complain they did not know it was an AI call This is a consent and communication issue, not a technical one — but it surfaces as a support escalation. The fix is upstream: ensure your application confirmation email tells candidates they may be contacted by an AI interviewer as part of the screening process. This also keeps you compliant with DPDP Act 2023 requirements around automated decision-making.

For a full view of BabbleBots' ATS integration capabilities, including how integrations work across different hiring volumes, see the product page.

Getting the Integration Live

The integrations above are not theoretical — they are live configurations running for enterprise teams in India today. The architecture is the same whether you're screening 50 candidates a week or 5,000.

What makes integration work in practice is alignment before configuration: knowing which stage triggers screening, what score means "proceed," and what data needs to live in the ATS for the downstream hiring team. The technical setup, once those decisions are made, is typically one working day.

If you're managing high-volume hiring across any of these three ATSes and want the AI screening layer to work without adding manual overhead, book a demo to see the integration in action with your specific ATS configuration.

FAQs

Q: Does BabbleBots integrate with Darwinbox? A: Yes. BabbleBots integrates with Darwinbox Recruit via webhook triggers and the Darwinbox REST API. Candidate data flows from Darwinbox to BabbleBots when they reach the configured hiring stage, and AI interview scores, summaries, and transcripts write back to the candidate profile in Darwinbox automatically. Setup takes 2–4 hours with API access in place.

Q: How long does ATS integration with an AI interview platform take? A: For Darwinbox, Keka, and Zoho Recruit, the integration is typically completed within one working day — assuming admin API access is available and the hiring stages have been pre-defined. The time splits roughly as: 1–2 hours for technical configuration, 30–60 minutes for testing with live candidates, and a brief alignment session to confirm score thresholds and routing logic.

Q: What data syncs between an AI interviewer and an ATS? A: The sync runs in both directions. From the ATS to BabbleBots: candidate name, mobile number, job ID, and the hiring stage. From BabbleBots back to the ATS: AI interview score, structured evaluation summary, recommended disposition (Proceed / Hold / Reject), a link to the full interview transcript, and the completion timestamp. The exact fields written back depend on your ATS configuration and which custom fields are available in your plan.

Q: Does BabbleBots integrate with Keka or Zoho Recruit? A: Yes to both. Keka Hire integration uses the Keka REST API with stage-based polling or a Zapier/Make intermediary for real-time triggering. Zoho Recruit integration uses OAuth API access combined with Zoho Blueprint for clean stage-transition triggers and score-based routing. Both integrations support full write-back of AI interview results to the candidate record.