— PROOF OF CONCEPT · AI RECRUITMENT

TalentIQ

Screen candidates with explainable AI.
AT A GLANCE
  • CATEGORYAI recruitment screening
  • ENGINELLM scoring + semantic
  • FORRecruitment agencies
  • STATUSProof of concept

Score, rank, and search your entire talent pool with deep LLM profile analysis — every decision explained in plain language, every weight recruiter-controlled. Search by intent, deliver branded shortlists, and run the full pipeline from job to hire. Built by IndiaNIC.

5
01 / SCORING DIMENSIONS
0–10
02 / EXPLAINED SCORE
300+
03 / RESUMES IN PARALLEL
<2 min
04 / TO SCREEN A BATCH
THE PROBLEM

Manual screening is slow, and it misses people.

Hours lost to resume review, keyword filters that drop qualified people, and black-box AI nobody trusts. TalentIQ answers all three — a demonstration of our NLP and AI/ML engineering.

~23h

spent reviewing resumes for a single hire

~75%

of qualified candidates missed by rigid keyword filters

0%

explanation from legacy black-box ranking AI

up to 3×

the role’s annual salary lost to one bad hire

EXPLAINABLE BY DESIGN

Every score, shown its work.

No mystery number. Each candidate gets a weighted breakdown, a fit verdict, a written rationale, and a standout insight you can actually defend to a client.

Priya R.
Senior React Engineer
Strong Match
8.4
/ 10
Skills Fit9.0 · 35%
Experience8.2 · 25%
Education7.5 · 15%
Keyword & Domain8.0 · 15%
Contextual7.8 · 10%
✦ AI Standout Insight

Shipped a design-system used across 6 teams — unusually strong platform ownership for the years of experience. Salary expectation within budget.

Illustrative scorecard — five weighted dimensions, a fit verdict, and a plain-language insight.
FOUR PILLARS

The whole agency workflow, in one system.

Score with transparency, search by meaning, collaborate with clients, and run the pipeline at scale.

  • 01 · PILLAR

    Explainable AI scoring

    Every candidate scored 0–10 across five configurable dimensions, with written rationale and a standout insight — no black box.

  • 02 · PILLAR

    Semantic talent search

    Plain-language search (“a senior React dev who led teams in fintech”) that understands intent, not keywords.

  • 03 · PILLAR

    Collaboration multiplier

    A branded, real-time client portal where companies review, comment on, and approve candidates — no email threads.

  • 04 · PILLAR

    Enterprise-scale pipeline

    Parallel resume processing plus a full Kanban hiring pipeline with stage logging and a complete audit trail.

HOW IT WORKS

Post, screen, collaborate.

From job posting to a branded client shortlist, the whole loop runs in one transparent system.

01

Post & upload

Create a job (or duplicate one), add candidates, and drop in resumes plus portfolio files and URLs.

02

AI screens in parallel

“Parse & Prepare” ingests everything, then the engine scores 300+ resumes simultaneously across five dimensions — and explains each.

03

Rank, search & collaborate

Review the ranked shortlist, search the talent pool in plain English, and hand clients a branded portal to comment and approve.

INSIDE THE SCORE

Transparent, tunable, recruiter-controlled.

The scoring engine keeps humans in charge — configurable weights, visible context, and rationale on every result.

  • 01 · FEATURE

    Five weighted dimensions

    Skills Fit, Experience, Education, Keyword & Domain Match, and Contextual Indicators — default weights overridable per job.

  • 02 · FEATURE

    Six-tier fit labels

    From Exceptional to Not Suitable, with admin-configurable score boundaries, names, and colours.

  • 03 · FEATURE

    Written rationale

    Key strengths, key gaps, and an overall recommendation in plain language for every candidate.

  • 04 · FEATURE

    AI Standout Insight

    A free-form observation beyond the grid — unusual expertise, self-directed learning signals — in a highlighted panel.

  • 05 · FEATURE

    AI Context Viewer

    See exactly what the model saw: job context, candidate sections, excluded items — read-only and fully transparent.

  • 06 · FEATURE

    Salary flag, never scored

    Compares expectation vs budget as a visual indicator only — it informs, it never sways the score.

CLIENT COLLABORATION

Hand clients a portal, not an email thread.

A branded, scoped space where clients review and approve in real time — and recruiters keep contact details and internal notes private.

Branded shortlist portal

Your brand, your credibility — clients review scores, fit labels, strengths, gaps, and standout insights in one place.

Real-time review & approval

Comment and approve with drag-and-drop stage management, replacing scattered email back-and-forth.

Scoped visibility

Clients see all AI result data — but never candidate contact details or recruiter internal notes.

Semantic talent pool

“React developer, rejected for location, score above 7” — parsed into filters plus vector ranking, with the interpretation shown.

BUILT RIGHT

A real build, not a mockup.

TalentIQ is a working proof of concept from our team. Want it shaped around your agency — or a different workflow entirely? Start with a rapid POC or browse the rest of the portfolio.

LLM scoring engineRAG semantic searchPostgreSQL + pgvectorReact + ViteNode.js + ExpressBullMQ + RedisGCP Cloud Run
— PROVE IT ON YOUR PIPELINE

Screening you can actually explain.

TalentIQ started as a proof of concept. Tell us your hiring workflow and the decisions you need to defend, and we'll stand up a working build on your data — fast.

— WHEREVER YOU ARE
hello@indianic.comWhatsApp Chat
RESPONSE TIME
< 4 hours
NDA
On request
FREE POC
3 – 5 days