Harsh Sharma

About Harsh Sharma

Full Stack Developer with 3+ years of experience building scalable backend systems, AI-driven applications, and data platforms in the financial technology domain. Currently working at ISS-STOXX where I design and develop high-performance microservices and AI-powered pipelines using Python, FastAPI, React, SQL, and distributed data systems. My work focuses on large-scale financial data processing, intelligent document analysis, and automation of regulatory workflows. I have hands-on experience with LLM applications, Retrieval-Augmented Generation (RAG), vector databases (FAISS), and semantic search to extract insights from complex financial documents such as SEC filings (10-K, DEF 14A) and EDGAR datasets. Key highlights of my work include: • Building fault-tolerant microservices handling 13M+ financial records using Python, MongoDB, Elasticsearch, and S3-compatible storage (MinIO) • Developing AI/ML pipelines with embeddings and vector search for contextual financial data analysis • Engineering scalable React frameworks and frontend architectures improving development speed by 90% • Optimizing large-scale SQL and data pipelines, improving query performance by 40% • Managing containerized microservices with Kubernetes and Docker for reliable deployments My core expertise lies in combining Software Engineering + AI/LLM systems + Data Engineering to build intelligent platforms that process and analyze massive financial datasets. Core Skills Python | FastAPI | React | Node.js | SQL | AI/LLM | RAG | Vector Databases | FAISS | Elasticsearch | Data Engineering | Microservices | Kubernetes | Docker | MongoDB | Kafka I’m always interested in working on AI-powered systems, LLM applications, and scalable data platforms.

Experience

1. Associate

   Iss Institutional Shareholder Services

   Jan 2026 — Present · Mumbai, IN

   Built a fault-tolerant Python microservice storing documents in MinIO (S3), using MongoDB (NoSQL) for flexible document metadata, MS SQL Server for relational metadata, and Elasticsearch for text-search,enabling scalable ingestion and fast search across 13M+ records.• Leading development of a configurable React framework built on Headless UI with JSON-driven UIcomposition and TanStack Query–based data orchestration (queries mutations), reducing time by 90% for newinterfaces and standardizing frontend architecture

Education

• SM SN shishu niketan school Jammu

  SSC, Science

• Government Hari singh Higher secondary school JAMMU

  HSC, Non Medical

• KJ Somaiya College of Engineering, Vidyavihar

  Bachelor of Technology, Information Technology

  2019 — 2023