Transaction Reference System
Implemented transaction-reference identification for RBI-compliant MOTO payment flows, making OTP and payment activity easier to trace.
↗ 50K+ monthly transactions · 70% fewer OTP-related escalations
I’m a Backend Engineer at ICICI Bank, working on systems across cards and payments.
My work spans Java/Spring Boot microservices, Python data pipelines, event-driven systems and payment infrastructure. Outside work, I build products around AI and financial technology.
Software Developer · Mumbai, India
Implemented transaction-reference identification for RBI-compliant MOTO payment flows, making OTP and payment activity easier to trace.
↗ 50K+ monthly transactions · 70% fewer OTP-related escalations
Built Python pipelines that process card statement and settlement data, enrich transactions with card metadata, and generate daily product-wise spend dashboards.
Built a Java Spring Boot microservice aggregating hedge-fund exchange-rate feeds via Kafka, with personalized rate computation based on customer relationships.
Replaced fixed GST payment pricing with credit-period-based fee computation using settlement and repayment dates.
↗ 30% growth in card spends over three months
Implemented AML compliance checks for high-throughput international transactions.
Backend Developer Intern
Designed Django REST APIs, integrated AWS EC2, S3 and RDS, and optimized database queries.
Backend Developer Intern
Built a Python/Django pipeline to ingest Excel data, extract structured records and update the database automatically.
From raw messages to financial clarity.
A cloud-native personal finance platform that transforms transaction messages into structured financial events and monthly spend insights. User-isolated, idempotent ingestion uses Gemini to extract merchants, transfers, refunds and amounts; a deterministic engine powers budgets, analytics and spend-threshold alerts.
AI-assisted product knowledge, deterministic recommendations.
Built a financial-product data and recommendation system covering 250+ credit cards. A multi-stage extraction pipeline normalizes product data; deterministic filtering and scoring evaluates rewards, eligibility, annual fees, spending patterns and card benefits.
Interested in backend engineering, distributed systems, payments and AI-enabled products.