AI-Powered Credit Scoring for Banks, Microfinance and BNPL companies
Mission: Revolutionizing credit scoring for the underserved segment in rural areas.
Introduction
A bank wants to issue a loan to a farmer, but without credit history, official income records, or digital financial traces, it has no proof of his ability to repay.
Rustam has:
Traditional System See:
Problem
Traditional credit assessment cannot evaluate borrowers who lack credit history, official income, and digital financial traces, leaving them underserved.
Non-systematic approach and Human factor
Manual decision-making leads to inconsistent evaluations and human bias
Limited and Low Quality Data
Lack of comprehensive credit history data reduces accuracy
Subjective and Unfair Decisions
Personal judgments create inequality in credit access
Outdated and Unchanged analyses
Static assessment methods fail to capture dynamic customer behavior
Slow Decision-making process
Lengthy evaluation cycles delay credit approval and customer service
Our AI-Powered Solution
MoliyAI uses AI-powered credit scoring to evaluate customers without formal income or credit history
Traditional Data
Alternative Data
Telecom Operators
Behavioral Data
ML MODELS
MOLIY AI RESULT
MoliyAI Platform
Experience intelligent credit assessment with real-time scoring and comprehensive risk analysis
Intuitive Assessment Interface
Streamlined credit evaluation workflow that combines traditional and alternative data for accurate decision-making
Real-time credit scoring with 87%+ accuracy
Comprehensive customer profile analysis
Instant approval decisions with detailed reasoning
Advanced Scoring Analysis
Visual credit score breakdown with transparent positive and negative factor analysis
Risk Factors
Identifies potential red flags
Positive Indicators
Highlights creditworthy signals
Real-Time Scoring
Get instant credit scores ranging from Poor to Excellent with detailed breakdown
Transparent Decision Making
Clear visualization of positive factors that support credit approval
Risk Assessment
Comprehensive analysis of negative factors to minimize NPL rates
How We Implement the Solution
AI/ML pipeline that processes alternative customer signals
Feature Engineering
Building 50–300 high-impact features, including:
- Debt ratios & income–obligation balance
- Payment discipline
- Transaction patterns
- Behavioral signals
- Regional & seasonal factors
Model Training
We train and benchmark multiple algorithms:
- LightGBM, XGBoost, CatBoost
- Neural Networks, Random Forest
- Logistic Regression, Ensembles
- Model selected by: F1-score
- Bad-client recall • AUC • Accuracy • NPL reduction
Explainability & Fairness
Ensuring transparent and unbiased decisions through:
- SHAP-based explanations
- Clear decision reasoning for credit officers
- Bias and fairness checks
- Regulatory compliance
- Audit trails and documentation
Our Roadmap
From MVP to Asian expansion
Current Stage
Launched
2025 — First National Startup MVP
CompletedTeam formation • Launch MVP • First BNPL pilot client
2026 — Scale Fergana Valley
CurrentDeploy Purchase Behavior • Deploy AIFS v1.0 • Onboard 10+ BNPLs
2027 — Cover All Regions
50+ clients across 12 regions • Custom AI modules for BNPL • Alternative credit signal partnerships
2028 — Asia Expansion
Pilots with Asian financial institutions
2029 — India, Indonesia, Pakistan
Localization • Offices • Train AI model with 10M+ records
Meet Our Team
Experts in AI, fintech, and alternative credit scoring
Bekhzod Aliev
CEO & Founder
- • MS in Business Analytics, University of Illinois (USA)
- • 6+ years of experience in Finance and Analytics
- • Business Analyst at Silon s.r.o.
- • Head of IT at Ehtirom Plus Microfinance
Otabek Davronbekov
CTO & Co-Founder
- • MS in Artificial Intelligence and Big Data, Anglia Ruskin University
- • 10+ years of experience in software development
- • Co-founder and ex-CTO of iTV Uzbekistan
Sevara Abdullaeva
Data Analyst
- • MS in Artificial Intelligence
- • 5+ years of experience in Data Analyst
- • Instructor at UNICEF Space Programming Tournament
- • Programming Lecturer at New Uzbekistan University
Jasur Bakhramov
ML Engineer
- • BS in Software Engineering
- • 7+ years of experience in AI and Software Development
- • Former ML Engineer at NEXIGN and iTV Uzbekistan
- • Finalist at MLC AI Hackathon
Yuliya Durova
Data Scientist
- • MS in Artificial Intelligence
- • 5+ years of experience in Data Science and Power BI
- • Programming Lecturer at New Uzbekistan University
Our Projects
Delivering cutting-edge AI and software solutions that transform how businesses operate across diverse industries
Spartak Jewelry
Uranus
Uranus Academy
Mercury Edu
The Redder
iTV
Hoopla
iTrack
Spartak Jewelry
Uranus
Uranus Academy
Mercury Edu
The Redder
iTV
Hoopla
iTrack
Our AI-powered credit scoring platform is deployed across various industries, helping businesses make better lending decisions and reduce NPL rates.
Awards & Recognition
Recognized by leading global organizations for innovation and excellence in FinTech
Global Startup Awards
Awarded in nomination of Best Fintech Startup project in Central Asia
Plug and Play Accelerator
Selected for global FinTech accelerator program
INMERGE Innovation Summit
Featured at leading innovation summit
Domino Ventures
Selected for BetterFuture Accelerator Program
Global Startup Awards
Awarded in nomination of Best Fintech Startup project in Central Asia
Plug and Play Accelerator
Selected for global FinTech accelerator program
INMERGE Innovation Summit
Featured at leading innovation summit
Domino Ventures
Selected for BetterFuture Accelerator Program
Global Startup Awards
Awarded in nomination of Best Fintech Startup project in Central Asia
Plug and Play Accelerator
Selected for global FinTech accelerator program
INMERGE Innovation Summit
Featured at leading innovation summit
Domino Ventures
Selected for BetterFuture Accelerator Program
These partnerships and awards validate our commitment to revolutionizing credit assessment through AI technology.