VDL Recruitment Intelligence Engine
Don't just source resumes. Engineer qualified hiring outcomes.
An AI + human recruitment process outsourcing (RPO) system that takes you from requisition to qualified candidate to successful joining — with a clear, evidence-backed reason behind every candidate we submit.
For enterprises, GCCs, IT services, SaaS companies and staffing firms that need faster, better-qualified hiring.
Why this candidate: 6 years building Spark pipelines on AWS in regulated payments data — directly matches the requisition's core must-haves.
- Spark + AWS, 5+ yearsEvidence: project history, certificationMatched
- Financial services domainEvidence: two payments employersMatched
- Team lead experienceMentors 2 engineers; no formal lead titlePartial
- Notice period ≤ 60 daysNot yet stated by candidateUnverified
- Compensation rangeMay sit above approved bandRisk
Recruiter validates: notice period, compensation expectation, interest in hybrid Noida role. Next: screen.
Illustrative “Why This Candidate?” brief. Every submission ships with one.
We don't present a candidate unless there is a credible reason the candidate fits the requirement.
Your recruiters and hiring managers receive “Why this candidate?” — not “Here is a resume.”
The same engine that finds buyers now finds hires.
The Recruitment Intelligence Engine is built on VDL's signal-led, evidence-backed GTM Revenue Intelligence Engine. The discipline stays the same. The object of intelligence moves from the account to the candidate.
Traditional recruitment
- Resume databases create volume.
- Keyword-only matching misses transferable skills.
- Bulk CV forwarding to hiring managers.
- One job board decides who gets seen.
- Recruiters search and screen manually.
- Pipeline visibility lives in spreadsheets.
VDL Recruitment Intelligence Engine
- Requisition intelligence creates relevance.
- Evidence-based matching, including transferable skills.
- Every submission carries a “Why this candidate?” brief.
- Multiple independent sources stack signals.
- Hermes orchestrates sourcing and research tools.
- ATS + control tower show every stage, SLA and bottleneck.
How a requisition becomes a qualified shortlist
Nine intelligence engines run in sequence before a single candidate is contacted.
Client intelligence
Who is hiring, for what, how many, where, by when, why — and what failed previously.
Requisition intake
Each job becomes a structured record: must-haves, good-to-haves, disqualifiers, success profile, urgency and difficulty.
Candidate persona
Target and alternative titles, adjacent and transferable skills, constraints and motivation indicators.
Talent market map
Where the talent exists: competitor and adjacent companies, clusters, certification pools, alumni and supply constraints.
Candidate discovery
14-step search across approved, lawful sources — your ATS, professional networks, job boards, portfolios and communities.
Candidate intelligence
What they've done, what they can do, what evidence supports it, what's unknown, and how to approach them.
Signal stacking
Twelve dimensions of fit, constraint and evidence quality combined into one explainable priority.
Explainable matching
Requirement-by-requirement comparison labelled matched, partial, unverified, mismatch or risk.
Candidate brief
The “Why this candidate?” brief, with recruiter questions and a recommended next action.
Signals are evidence to investigate — never assumptions.
Career and talent signals we track
A signal does not mean a candidate is available. High priority requires strong requirement fit, strong evidence, relevant experience, acceptable constraints, credible availability and appropriate seniority — together.
Every requirement gets a label
- ATechnical fit
- FCompensation fit
- BRelevant experience
- GAvailability
- CIndustry fit
- HCareer alignment
- DSeniority fit
- IRequirement coverage
- ELocation fit
- JEvidence strength
The score prioritizes recruiter work. It never replaces recruiter judgment.
Hermes orchestrates. Recruiters decide.
Hermes is the RPO brain. It doesn't replace your recruitment tools — it decides which tool to call, in what order, with what question, and combines the evidence into one candidate record. Composio MCP connects it to your ATS, HRIS, calendars and communication tools.
A recruitment lead can simply ask:
For the Senior Data Engineer requisition, map the talent market, find 40 candidates with verified Spark and cloud experience, apply disqualifiers, score them, draft briefs for the top 10, and place them in the ATS for recruiter review.
14-layer architecture
- RPO brainHermes
- ReasoningClaude or other approved model
- ConnectivityComposio MCP
- Talent dataCandidate databases, networks, job boards
- ResearchExa, web research, approved sources
- Company intelligenceCareer pages, hiring activity, tech data
- ATSSystem of record
- HR / client systemsHRIS, CRM, client systems
- EngagementEmail, LinkedIn, messaging, phone
- ScreeningAI-assisted + human screening
- InterviewCalendar, scheduling, feedback
- OfferApprovals, negotiation tracking
- AnalyticsRPO, recruiter and client dashboards
- GovernancePrivacy, compliance, audit, oversight
AI does
- Research
- Search
- Enrichment
- Matching
- Classification
- Summarization
- Drafting
- Scheduling support
- Follow-up
- Data entry
- Reporting
- Workflow orchestration
Your recruiter does
- Requirement validation
- Candidate relationship
- Nuanced assessment
- Final screening
- Sensitive conversations
- Candidate counseling
- Hiring-manager relationship
- Exception handling
- Offer negotiation
- Hiring recommendation
- Rejection communication
- Escalations
You decide
- Define the hiring need
- Validate requirements
- Conduct interviews
- Provide feedback
- Make hiring decisions
- Approve offers
- Own the employment decision
RPO doesn't end at the offer letter.
We run engagement, screening, interviews, offers and joining — and track every hire to 90 days.
Candidate outreach
Signal → role relevance → opportunity → relevance to the candidate → proof → question. Across email, LinkedIn, phone and — where appropriate — SMS or WhatsApp, always respecting opt-outs, platform rules and privacy.
Interview intelligence
Hermes flags interviews with missing feedback, candidates waiting too long, hiring-manager delays, disengagement risk, conflicting feedback and repeated failures — before they cost you the hire.
Offer and joining
Offer date, amount, negotiation, acceptance or decline reason, notice period, documentation, preboarding, day-one, and 30/60/90-day status — because the outcome is a successful joining.
Volume
- Open and new requisitions
- Sourced, contacted, responded
- Qualified and screened
- Submissions and interviews
- Offers and joinings
Velocity
- Time to source
- Time to submit and interview
- Time to offer and fill
- Hiring-manager delay
- SLA status with escalation
Quality
- Interview-to-submission rate
- Offer acceptance
- Candidate drop-off
- Joining failure
- Quality of hire
See the full 33-step end-to-end workflow
- Client discovery
- Client intelligence
- Requisition intake
- Job intelligence
- Candidate persona engineering
- Talent market intelligence
- Candidate discovery
- Candidate enrichment
- Candidate verification
- Signal detection
- Candidate matching
- Candidate scoring
- Candidate brief
- Recruiter human review
- Candidate outreach
- Candidate response
- AI-assisted pre-screening
- Human recruiter validation
- Candidate submission
- Client review
- Interview scheduling
- Interview
- Interview feedback
- Next-round decision
- Offer
- Negotiation
- Offer acceptance
- Preboarding
- Joining
- 30/60/90-day tracking
- Quality-of-hire feedback
- Learning loop
- Improve the next cycle
Hiring decisions affect livelihoods. Governance is built in.
Every high-impact decision has human oversight and an audit trail. And every outcome — responses, interviews, declines, joinings, quality of hire — feeds a learning loop that improves search, scoring, outreach and screening for the next requisition.
What we track
- Candidate consent where required
- Communication preferences and opt-outs
- Data source, freshness and retention
- Access control and audit trail
- AI-generated outputs and human approvals
- Client policies, employment and privacy requirements
What AI is never allowed to do
- Make final employment decisionsRecruiter recommends, client decides
- Infer protected characteristicsExcluded from persona, matching, scoring
- Use irrelevant personal traitsJob-related dimensions only
- Treat unverified data as factVerified / stated / unknown on every field
- Auto-reject on an opaque scoreExplainable scorecard + human review
Live in 90 days: build, test, scale.
Client and requisition intake, candidate persona, ATS integration, candidate data connections, talent market intelligence, Hermes workflows, scoring, briefs, recruiter dashboard, compliance controls.
Output: working engine, first requisitions, first qualified candidates.
Sourcing, matching, signal quality, scoring, outreach, screening, recruiter and interview workflows, hiring-manager feedback, ATS automation.
Output: validated workflow, sources, matching and outreach.
More requisitions, recruiters and candidates; multichannel engagement; automation; SLA monitoring; client dashboards; talent-pool reuse; learning loops.
Output: a repeatable AI-assisted RPO system.
Judge us on your requisitions, not our database size.
Before you trust the system at scale, we run it on a defined set of your real open roles. You see the research, the shortlist and the reasoning.
- Requisition intelligence record for each role
- Candidate persona with transferable skills
- Talent market map and supply constraints
- Discovered, enriched and verified candidates
- Scored shortlist with explainable scorecards
- “Why This Candidate?” briefs for top candidates
- Recruiter review of the shortlist
- Live demo: outreach, screening, ATS update, dashboard
Engagement models that fit your hiring volume.
Chosen by hiring volume, role complexity, geography, seniority, number of recruiters, SLA, technology needs and how involved your team wants to be.
Business KPIs
- Requisitions filled
- Time-to-fill
- Offer acceptance
- Joining rate
- Cost per hire
Candidate KPIs
- Response rate
- Qualified rate
- Submission rate
- Interview rate
- Acceptance rate
Recruiter KPIs
- Requisitions per recruiter
- Screens completed
- Submissions and joins
- Productivity
- SLA compliance
Client KPIs
- Hiring-manager response
- Feedback turnaround
- Candidate acceptance
- Requisition aging
- Bottlenecks
Built by VDL — Viral Digital Labs.
VDL applies signal-led, evidence-backed engineering to the two pipelines every growing company depends on: revenue and talent. The Recruitment Intelligence Engine extends our GTM engineering discipline into recruitment process outsourcing.
VDL is led by Abhishek Sharan, Enterprise Revenue Transformation Consultant and former AVP Sales (North America) at Newgen Software — 25+ years in enterprise sales across Fortune 500, BFSI, ISV and technology companies, with hands-on experience building AI-enabled systems using Clay, Claude, Apollo, Composio and automation.
- We don't sell resume databases; we engineer qualified hiring outcomes.
- Requisition intelligence comes before sourcing.
- Evidence is attached to the candidate before submission.
- Recruiters own judgment. Clients own hiring decisions.
- The system compounds through feedback.
Tell us what you're hiring for.
Share a few open roles. We'll come back with a proof-of-work plan: which requisitions to run, what the talent market looks like, and what a shortlist from us would include.