Step 1 of 7
Guided Step Title
This is the step description.
πŸ“‹ JD Upload N/A | 50ms
β†’
🧠 JD Understanding N/A | 1.2s
β†’
πŸ” Semantic Retrieval 100K βž” 1K | 142ms
β†’
βœ… Truth Engine 1K βž” 100 | 4.5s
β†’
πŸš€ FutureFit Engine 100 | 2.1s
β†’
πŸ† Ranking Engine 100 | 80ms
β†’
πŸ“€ Top 100 Generation 100 | 30ms
India Runs 2026 Β· Track 1 Β· All 9 Modules Operational

The Future of Hiring is
Human Potential Intelligence

TalentLens X doesn't match resumes β€” it discovers human potential. Our 9-module AI pipeline semantically retrieves, truth-validates, and future-predicts every candidate, delivering fully explainable ranked results in under 3 minutes for 100,000 candidates.

9-Module AI Pipeline Architecture
Module 01
πŸ“‹
JD Understanding Engine
Module 02
⚑
Skill Extraction
Module 03
🎯
Embedding Generation
Module 04
πŸ”
Semantic Retrieval
Module 05
βœ…
Truth Engine
Module 06
πŸš€
FutureFit Engine
Module 07
πŸ†
Ranking Engine
Module 08
πŸ“Š
Recruiter Dashboard
Module 09
πŸ“ˆ
Visualization & Insights
100K
Candidates Processed
9
AI Modules Active
<3min
Full Pipeline Runtime
100%
Explainable Rankings
πŸ‘₯ Developer & Contact Profile
Ruthvik Goud β€” Sole Developer & AI Architect
GitHub: @ruthvikgoud16  Β·  LinkedIn: Ruthvik Goud Profile  Β·  Email: bathiniruthvik370@gmail.com
Developed for the Redrob Hackathon. Special thanks to the Redrob team and mentors for their guidance, test dataset, and evaluation schema.
Module 01 β€” 02
JD Understanding Engine

AI-powered job description parser. Extracts structured requirements, skills, and context for semantic matching.

πŸ“‹ Job Description Input
Senior AI Engineer Β· Redrob AI
⚑ AI Processing Pipeline
πŸ“NLP Tokenization & PreprocessingPENDING
⚑Skill Entity Extraction (NER)PENDING
🧠Intent Classification & Role ParsingPENDING
🎯JD Embedding Generation (BGE-Large-1024d)PENDING
πŸ“ŠStructured JSON Output GenerationPENDING
Module 03 β€” 04
Semantic Retrieval Engine

Vector similarity search across 100,000 candidates. No keyword matching β€” pure semantic understanding via 1024-dimensional embeddings.

πŸ”§ Search Configuration
Embedding Model
BGE-Large-EN-v1.5
Vector Dimensions
1024-dim HNSW Index
Similarity Metric
Cosine Similarity
Strategy
Dense + BM25 Hybrid
Search Latency
142ms end-to-end
Search Progress
Awaiting JD embedding...
0 / 100,000 scanned
Candidates scanned100,000
Pool retrieved0
Vector search latency142ms
🎯 Embedding Space (t-SNE 2D projection)1024Dβ†’2D
πŸ“₯ Retrieved Candidate Pool
0 retrieved
Module 05 β€” 08
Recruiter Intelligence Dashboard

Truth Engine + FutureFit Engine + Explainable Ranking. Every score backed by evidence. No black-box outputs.

Final Score Formula:
SemanticΓ—40%+ TruthΓ—25%+ FutureFitΓ—20%+ ExperienceΓ—15%
Weights calibrated on recruiter engagement data
Quick Filters
All Candidates 20
Top 5 5
Top 10 10
Truth Score β‰₯85 8
FutureFit β‰₯88 11
Notice ≀30 days 14
Location
Bengaluru 8
Hyderabad 5
Pune 4
Delhi NCR 3
AI Domain
Info. Retrieval 12
NLP Engineering 8
Recommendation 6
Search Systems 9
Showing 20 candidates Β· Ranked by Final Score
CANDIDATE
SEMANTIC
TRUTH
FUTUREFIT
EXP.
FINAL
Module 06 β€” Explainable Ranking Engine
Ranking Explainability

Every ranking decision mapped to specific evidence. No black boxes. Each score traced to source data.

Explaining rank for:
Module 09 β€” Visualization & Insights
Hiring Intelligence Analytics

Aggregated insights across the candidate pool. Identify patterns, skill gaps, and hiring opportunities.

TalentLens X β€” Interactive Pipeline Showcase
Automated Demo Mode

Experience the 3-minute end-to-end Candidate Discovery, Verification, and Ranking pipeline.

Visual Workflow Tracker
1
JD Upload
Awaiting...⏱ ~50ms
↓
2
JD Understanding
Awaiting...⏱ ~1.2s
↓
3
Semantic Retrieval
Pool: 100K βž” 1K⏱ ~142ms
↓
4
Truth Engine
Audit: 1K βž” 100⏱ ~4.5s
↓
5
FutureFit Engine
Score: 100⏱ ~2.1s
↓
6
Ranking Engine
Rank: 100⏱ ~80ms
↓
7
Explain #1 vs #2
Evidence Audit⏱ ~1.5s
↓
8
Export Top 100
Save CSV⏱ ~30ms
Stage 1: Awaiting Init IDLE
⚑

TalentLens X Demo Protocol

Click the button below to initialize the automated end-to-end recruitment pipeline demo, scanning 100,000 candidates.