Jayam Gupta

As a Data Science Consultant with hands-on experience in building real-world AI systems, I specialize in developing scalable, intelligent solutions that bridge the gap between unstructured data and business insight. My journey spans across healthcare and insurance domains, where I’ve led end-to-end projects involving LLM integration, backend development, and ETL pipelines under high-pressure, production-grade environments.

At Consint.ai, I revolutionized fraud detection for CARE Health Insurance by delivering an automated pipeline that processed 200–300 insurance policy documents daily, reducing human intervention and achieving sub-100ms response times. I designed custom entity extraction algorithms using a local Qwen-7B LLM and optimized them for vendor-specific templates, achieving an 85% improvement in accuracy. I also authored 20+ fraud detection rules and architected a dynamic rule engine that supports rapid updates and integrations.

Previously at Aidwise.ai, I took complete ownership of backend development projects, quickly mastering FastAPI, MongoDB, and GitHub Actions. I reverse-engineered complex clinical data pipelines from notebooks into robust ETL systems and built high-accuracy algorithms to interface with the National Drug Code API. My LLM-focused contributions consistently turned ambiguous ideas into production-ready modules and earned recognition for speed and precision.

My technical arsenal includes Python, SQL, FastAPI, Docker, and MongoDB, with strong foundations in CI/CD, REST API development, and prompt engineering. I have implemented parallelized ML clustering for CKD diagnosis, extracted structured summaries from 30,000+ unstructured patient visits, and built an intelligent chatbot capable of responding to NL queries with visual data insights in real time.

Certified in Machine Learning (Udemy), DevOps, and Cloud Computing (IBM), I combine academic grounding with practical know-how. I hold a B.Tech in Information Technology from KIIT University, along with a minor in Financial Economics.

I’m passionate about building systems that are fast, reliable, and deeply aligned with user needs. Whether it’s transforming clinical workflows or designing fraud rule engines, I bring a balance of analytical rigor and software craftsmanship to every problem I solve.

Explore my open-source work on GitHub github.com/jayam, published Python packages on PyPI (pypi.org/project/WrappedLLM), and projects at portfolio.guptajayam.com. I’m always open to solving tough problems, working with purpose-driven teams, and pushing the limits of applied AI.