The AI Lending Platform Transformation A Fintech Success Story Meant to Scale

A Fintech Success Story That Scaled When Everything Was on the Line

The Moment of Truth

In early 2024, AI Lending Platform — India’s fastest-rising AI lending platform was staring at disaster weeks before launch.
“We expect 4,000+ concurrent users and thousands of real-time AI credit decisions per minute,” the CTO said. “Our on-premise MySQL is already dying at 500 ms latency. One outage and we lose customers, trust, and RBI compliance.”

The Harsh Reality We Discovered

Legacy Windows servers + bare-metal MySQL with zero failover Queries routinely >500 ms (unacceptable when conversion drops 1% per 100 ms) No encryption-at-rest strategy for borrower PII Windows-centric enterprise teams with deep Active Directory, SQL Server, and PowerShell dependencies Zero chance of surviving launch traffic

The New Foundation: Full-Stack Windows-Integrated AWS Architecture

We delivered a completely cloud-native, RBI-compliant platform that feels 100% native to Microsoft enterprise environments while delivering open-source performance and cost.

Core Database Layer (Dual-Stack Ready)

Primary production: Amazon RDS for PostgreSQL Multi-AZ (chosen for cost + performance) Identical architecture pre-validated and ready for Amazon04 RDS for SQL Server or SQL Server Always-On on EC2 Windows 200 tables, 140 constraints, 60 indexes — fully compatible with SQL Server data types and T-SQL patterns p95 latency dropped from 500+ ms → <20 ms at 4,000 concurrent connections

Deep Microsoft Windows & Enterprise Integrations (The Real “Windows Part”)

Microsoft Windows-Based Compute

60% of ECS cluster runs Windows Server 2022 containers (AMI optimized)  Critical .NET Framework 4.8 and .NET 6/8 microservices (loan scoring, compliance rules engine) run natively on Windows nodes Zero code changes for AI Lending Platform’s legacy Windows services — simply containerized and deployed

Microsoft Active Directory — The Single Source of Truth

On-premises Active Directory federated via AWS Managed Microsoft AD + IAM Identity Center (AWS SSO)
All developers, DevOps, and finance teams log in to AWS Console, CLI, and internal tools using their existing AD credentials
Group-based permissions (e.g., “LoanOps-Admin”, “Compliance-Audit”) mapped directly to IAM policies

Microsoft SQL Server Integration (Live & Ready)

Shadow Amazon RDS for SQL Server instance kept in sync via DMS CDC for downstream Windows-only reporting tools (SSRS, Power BI DirectQuery, legacy .NET apps)
All encryption, HA, and backup patterns identical to production PostgreSQL
AI Lending Platform can flip any new module to SQL Server tomorrow without re-architecting

Microsoft PowerShell & .NET-Powered Automation (The Team Never Left Their Comfort Zone)

Entire IaC secondary layer written in PowerShell DSC + AWS Tools for PowerShell (alongside Terraform)
GitHub Actions runners (Windows-latest) execute PowerShell/.NET scripts for:
Secret rotation
Compliance report generation
Windows patch compliance validation
Automated failover testing
Zero training required — AI Lending Platform’s Windows admins became cloud admins overnight

AWS Systems Manager — Windows First-Class Citizen

Patch Manager automatically patches Windows containers and EC2 Windows instances
Session Manager replaces RDP/SSM with audited, bastion-less access
Inventory + State Manager enforce golden Windows AMIs across the fleet
Run Command executes PowerShell remediation scripts in <30 seconds fleet-wide

Encrypted Windows Storage

Amazon EBS gp3 volumes (KMS-encrypted) attached to Windows ECS tasks for:
High-speed local SSD caching of KYC documents
Windows event logs & application tracing
Temporary .NET memory dumps (auto-encrypted)

Real-Time Processing & Scale

Mixed Linux/Windows ECS cluster with auto-scaling to 4,500+ connections
ElastiCache for Redis (Windows services use StackExchange.Redis natively)
CloudFront + Route 53 delivering sub-second responses India-wideleo.

Security & RBI Compliance (Microsoft-Style Governance)

AWS KMS + Secrets Manager (90-day rotation via PowerShell Lambda)
All Windows instances domain-joined to AWS Managed Microsoft AD
CloudTrail + GuardDuty + quarterly IAM/AD access reviews
Hardware MFA + privileged access logging for all root and AD admin actions

Zero-Downtime Migration

AWS DMS + SCT converted MySQL → PostgreSQL
Parallel DMS CDC pipeline kept SQL Server shadow instance in sync
Blue/green Windows + Linux ECS deployments → <12 minutes total cutover
Tested fallback to on-premise Windows/SQL Server in read-only mode

Architecture Diagram

Launch Day: The Windows Stack Didn’t Even Break a Sweat

4,300+ concurrent users at peak
Windows-based .NET scoring services processed 180,000+ AI decisions/hour
p95 query latency: 18 ms
End-to-end offer generation: <90 ms
Uptime: 99.999% (one brief Multi-AZ failover — 6 seconds)
Zero customer impact

Beyond Survival — The Wins No One Expected

50% TCO reduction vs on-premise Windows + SQL Server licensing
Windows patch compliance went from 63% → 100% automatically
PowerShell-driven monthly cost & security reports now take 4 minutes instead of 4 days
AI Lending Platform’s Windows team now deploys production changes faster than their Linux team