Investment Tracker Platform
NAV Fund Administration Group
Led development of a comprehensive Investment Tracker platform integrating Document Management System, external APIs, and internal accounting systems. Built Spring Boot schedulers and React dashboards that reduced document processing delays by 40% and improved transparency across 5,000+ funds.
Investment Tracker Platform
NAV Fund Administration Group
Led development of a comprehensive Investment Tracker platform integrating Document Management System, external APIs, and internal accounting systems. Built Spring Boot schedulers and React dashboards that reduced document processing delays by 40% and improved transparency across 5,000+ funds.
Cassandra Reporting Database
NAV Fund Administration Group
Designed and deployed a Cassandra-based reporting database to offload SQL Server workloads, supporting 10M+ records with optimized compaction strategies for write-heavy analytics.
Kafka Data Ingestion Pipeline
NAV Fund Administration Group
Developed a Kafka-based data ingestion pipeline replacing manual batch imports, with staging and reconciliation layers ensuring 99.9% reliability and reducing processing latency.
Client Payments Settlement Module
NAV Fund Administration Group
Created a comprehensive Client Payments Settlement Module with Aging Tracking and automated invoicing/notifications using Java microservices architecture.
5TB SQL Server Database Partitioning
NAV Fund Administration Group
Spearheaded database partitioning of a high-traffic 5TB SQL Server database, reducing query response time by 40% and streamlining management of millions of records.
Automated Tax Filing Solution
NAV Fund Administration Group
Delivered an automated tax filing solution integrating ReactJS frontend, Java Spring Boot backend, and SQL Server database, reducing filing cycle time by 50%.
Tag-Based Reporting Framework
NAV Fund Administration Group
Engineered a tag-based reporting framework unifying 20+ database tables, enabling self-service reports and reducing IT requests by 35%.
AI Tools Integration for Developer Productivity
NAV Fund Administration Group
Integrated AI-powered tools (Cursor, Code Ant AI) into development workflows, streamlining code reviews and reducing repetitive boilerplate coding. Established AI-assisted coding practices that improved code quality consistency while reducing review turnaround times.
Custom Grouping System
NAV Fund Administration Group
Led the development of a real-time data grouping system across Java and .NET services, leveraging parallelism and multithreading to dynamically aggregate large data for consolidated reporting. Developed robust business logic and comprehensive error handling mechanisms.
Kerberos Authentication Implementation
NAV Fund Administration Group
Engineered and deployed Kerberos authentication across multiple services, significantly enhancing system security. Conducted rigorous testing and troubleshooting to achieve seamless Kerberos integration across diverse environments and services.
Java 8 → 21 Microservices Migration
NAV Fund Administration Group
Migrated 5+ production microservices from Java 8 to Java 21 (LTS), ensuring long-term platform supportability. Leveraged modern garbage collectors (G1/ZGC) and runtime enhancements to optimize heap utilization and reduce GC pauses.
Redis Cluster Migration
NAV Fund Administration Group
Orchestrated migration of Redis to a clustered setup, enabling fault tolerance and distributed caching. Achieved 66% higher throughput and optimized memory utilization in high-volume ingestion pipelines. Improved system resilience and stability under peak workloads.
LeetCode Problem Solving Journey
Personal Development
Consistently solving algorithmic problems to strengthen problem-solving skills, with 157 problems solved across various difficulty levels and topics.
UAV Image Instance Segmentation
North Eastern Space Applications Centre
Trained Mask R-CNN for 3-class instance segmentation of UAV image data, achieving 0.46 average accuracy. Annotated images using VGG Image Annotator and implemented semi-supervised U-net model for flood water detection.
Real-time Depth Estimation
Inspired Automation Future Technologies
Calibrated stereo cameras and designed real-time object depth estimation algorithm using HSV values. Implemented pathfinding algorithm for robot navigation in grid environments.
Automated Attendance System
IIT Gandhinagar - Prof. Nipun Batra
Designed real-time automated attendance system using image processing. Implemented and compared Eigen faces, Fisher faces, LBP faces, and Siamese networks for face recognition.
Sampling to Speed Up Clustering
IIT Gandhinagar - Prof. Anirban Dasgupta
Implemented Lightweight coreset construction algorithm and compared performance metrics of various clustering algorithms with and without lightweight coresets using KDD dataset.
16-Bit Bootloader
IIT Gandhinagar - Prof. Nipun Batra
Built a 16-bit bootloader in assembly language following "Writing a simple operating system - from scratch" book. Implemented low-level system programming concepts and OS fundamentals.
Image Denoising using Wavelet Shrinkage
IIT Gandhinagar - Prof. Shanmuganathan Raman
Studied and implemented wavelet shrinkage algorithm to denoise RGB images. Applied advanced signal processing techniques for image enhancement and noise reduction.
Future Frame Prediction
IIT Gandhinagar - Prof. Shanmuganathan Raman
Implemented deep learning models for predicting future frames in video sequences using temporal information and convolutional neural networks.