Nishant Prakash

Visa Developer Platform (VDP)

• Designed and implemented a runtime microservice from scratch to perform encryption and decryption of sensitive payloads in transit, handling over 5 million requests daily, and significantly enhancing data security and regulatory compliance.
• Integrated Model Context Protocol (MCP) with Visa Developer Platform Gateway, exposing core gateway services (AuthN, ARM, NPPS) as standardized AI-accessible tools — enabling intelligent observability, debugging, and developer self-service.
• Engineered a scalable solution to support large payloads exceeding 10GB in API requests, enabling seamless ingestion of high-volume data from external partners without performance degradation.
• Migrated legacy file-based logging system to a Kafka-based distributed logging pipeline, improving log throughput by 40% and enabling real-time analysis across gateway services.
• Led the complete Splunk to Humio migration initiative, successfully transitioning over 100+ dashboards and 200+ alerts, reducing log analysis latency by 35% and cutting monitoring costs through licensing optimization.

Finexperience – Digital Mortgage Platform

• Designed and developed a loan management web application tailored for loan officers to view, track, and manage borrower profiles and loan details, reducing manual processing time by 40%.
• Implemented key modules for document upload, e-signature, and e-disclosure using Spring Boot, OAuth, and AWS services, enabling a paperless mortgage approval process across 100+ banking clients.
• Built a flexible system to connect with different financial data sources using REST APIs and Kafka, which reduced the time to onboard new data providers by 50%.
• Collaborated with cross-functional teams including product and QA to ensure successful delivery of microservices with 99.9% uptime and compliance.

• Built a real-time transaction risk engine for a fintech client using Java and Kafka, capable of processing 10K+ transactions/sec, improving fraud detection accuracy and reducing false positives by 30%.
• Led a data migration project for a healthcare client, moving 5+ TB of patient data to Snowflake using Informatica, and built Tableau dashboards to visualize KPIs like claim rejections and patient discharge trends.
• Performed EDA using Python (Pandas, NumPy, Matplotlib) to uncover high-risk patient segments, contributing to a 10% reduction in re-admission penalties and improved clinical decision-making.