Jeet Sharma
Backend Engineer • Distributed Systems • Production AI
Building scalable infrastructure and data-intensive platforms from zero to production. Currently @ UMass Amherst pursuing MS in Computer Science.
Building scalable infrastructure and data-intensive platforms from zero to production. Currently @ UMass Amherst pursuing MS in Computer Science.
Founding engineer building production-grade distributed systems and AI platforms.
From zero to enterprise scale.
Production-grade digital forensics tool for processing and analyzing digital evidence. Enterprise platform handling large-scale forensic investigations with advanced data extraction and analysis capabilities.
Designed and deployed an extensive end-to-end pharmaceutical supply chain solution leveraging blockchain technology, ensuring tamper-proof tracking of medical drugs and presented the solution to the Indian Medical Association.
Fine-tuned DistilBERT and CodeBERT classifiers to dynamically route LLM requests, reducing inference costs by ~70% while preserving output quality. Implemented guardrails and fallback mechanisms for predictable system behavior.
Built a full-stack AI-driven recipe generation platform enabling users to create personalized meals based on available ingredients, dietary preferences, and allergies, leading to a 40% reduction in recipe search time.
Developed a web app for Covid-19 detection from Chest X-ray images, achieving 90%+ accuracy in classification. Improved model performance by 30% using data augmentation and hyperparameter tuning.
University of Massachusetts Amherst
Sep 2024 – May 2026
GPA: 3.86/4.0
University of Mumbai
Aug 2019 – May 2023
GPA: 3.87/4.0
May 2025 – Aug 2025
Built multi-tenant RAG backend for business customer support across chat and messaging platforms with semantic retrieval using PostgreSQL pgvector for grounded LLM responses
May 2023 – Aug 2024
Architected AI-driven forensics platform from zero to production serving enterprise clients
Aug 2022 – Apr 2023
Built distributed graph database with sharded architecture and real-time streaming pipeline