Headline

AI/ML Engineer | Machine Learning | Exp: Insurance, Banking, Supply Chain, Mobility & IT

About

With 4+ years of experience, I build production-grade AI, Machine Learning, NLP, and Generative AI solutions that solve complex business problems across Insurance, Banking, Supply Chain, Mobility, and IT. My work has delivered tangible results—from protecting ~$2M in revenue through predictive AI, to reducing compliance screening effort by 87%, and accelerating contract review by 78% across 10M+ legal documents. I specialize in taking AI from experimentation to scalable, enterprise-ready deployment, combining data science, MLOps, and modern GenAI frameworks to create solutions that drive efficiency, reduce risk, and unlock business value.

Location

Gurugram

AI stack

antigravity

codex

Work experience

AI / ML Engineer at Optum, Jul 2026 - Present · 2 mos, Gurugram

Data Specialist at Marsh, Oct 2024 - Jun 2026 · 1 yr 9 mos, Designing AI-powered compliance systems at one of the world’s largest insurance and risk advisory firms. → Cut manual compliance screening by 87% — Power BI dashboards over unified Salesforce + SharePoint + Teradata data across 500+ client portfolios → Deployed legal risk classifier over 10M+ contracts — 78% faster contract review (AUC 0.84, F1 0.80) → Built RAG agent (LangChain + FAISS + GPT) over 2M+ legal documents — 60% faster clause lookup for legal teams → Engineered automated client risk scoring model (AUC 0.81, Precision 82%) with plain-English compliance narratives for leadership → Designed agentic workflows for contract summarisation and risk flagging, piped into Power Automate for real-time alerts Azure Databricks • PySpark • LangChain • spaCy • Power BI • Azure SQL

Data Scientist at FedEx Express, Dec 2022 - Sep 2024 · 1 yr 10 mos, Gurugram, Haryana, India, Delivered end-to-end data science, predictive modeling, and executive reporting as a Data Scientist supporting MEA and MEISA markets — one of FedEx’s highest-volume international trade regions. Data Engineering & BI → Built and maintained PySpark ETL pipelines on Azure Databricks processing 100M+ daily transactions; optimized partitioning strategy to cut analytics latency by 60% for near real-time business access → Automated Power BI and Tableau revenue dashboards for senior leadership across two regions; eliminated manual weekly refresh cycles, cutting maintenance effort by 50% Machine Learning & Predictive Analytics → Built XGBoost churn prediction model (88–92% accuracy) with walk-forward validation; retention strategies driven by model outputs reduced customer attrition by 12%, protecting ~$2M in revenue → Engineered 45 features on 134M shipment records across 50+ trade corridors; LightGBM demand forecast hit 11% MAPE — 39% better than SARIMA — giving supply chain teams reliable forward visibility Stack: Python • PySpark • Azure Databricks • XGBoost • LightGBM • Power BI • Tableau • SQL

Data Analyst at Proziod Analytics Pvt. Ltd., Sep 2020 - Jul 2021 · 11 mos, Bengaluru, Karnataka, India, Identify, analyze, and extracting large sets of data, data cleansing, and modeling data with the purpose of understanding or making conclusions from the data for decision-making purposes and presenting the data using SQL queries and visualization tools such as Power BI. Identify, develop and implement statistical techniques that address business challenges and add value to the organization. Working closely with the development team to provide an integrated solution for platform enhancements and building and manage dashboards.

Data Analyst at MapUp, Jan 2020 - Jun 2020 · 6 mos, Gurgaon, Haryana, India, Used BeautifulSoup and selenium to extract desired data from a different website. Performed data manipulation, transformation, and cleansing. Designed, recorded, and executed macros to automate data entry inputs. Utilized SQL,Power Bi, Tableau and other dashboard Data visualization tool sets for Data intelligence and analy

Education

Bachelor of Technology - BTech, Information Technology, SRM IST Chennai