9 years of experience in Machine Learning, Deep Learning and NLP, LLM. Working with Anheuser - Busch InBev as Data Scientist.
Leading development of Cash.AI, a GenAI-powered document intelligence platform using LLMs to automate remittance and claims extraction across heterogeneous customer formats (Excel, PDF, scanned images).
Designed a scalable, country-agnostic processing pipeline, reducing remittance processing time to ~5 minutes per batch and enabling near real-time finance operations.
Deployed the product across seven countries, achieving enhanced data accuracy, ~70% increase in analyst efficiency, and an 82% overall touchless processing rate, driving measurable improvements in OTC performance KPIs.
Ensured production-grade security and governance by remediating vulnerabilities via Snyk, SonarCloud, and Apiiro, and completing MICS (Minimum Internal Control Systems) compliance for GenAI deployment in a regulated environment.
Developed a weekly price direction prediction model using ensemble techniques (Random Forest, XGBoost), achieving 77% F1-score and delivering $1.2M in cost savings.
Implemented Optuna-based hyperparameter optimization to improve robustness and generalization.
Built a 12-week price forecasting model using Orbit DLT, achieving 76% directional accuracy and generating $2.3M in savings for procurement hedging decisions.
Created an ensemble model for hedging instrument selection (forwards/options), achieving an F1 score of 88% and delivering $1.6M in savings.
Leading development of Cash.AI, a GenAI-powered document intelligence platform using LLMs to automate remittance and claims extraction across heterogeneous customer formats, Designed a scalable, country-agnostic processing pipeline, reducing remittance processing time to ~5 minutes per batch, Deployed the product across seven countries, achieving enhanced data accuracy and increased analyst efficiency, Ensured production-grade security and governance for GenAI deployment in a regulated environment
Developed a real-time sentiment analysis tool, reducing manual effort by 80% and enhancing Back Office efficiency.
Implemented a clustering model for risk assessment, categorizing over 50k companies to support CMS, EWS, and E-Adjudication.
Developed a real-time sentiment analysis tool, Implemented a clustering model for risk assessment
Automated KYC document extraction, reducing processing time by 60%.
Built a fraud detection model for credit card transactions, supporting risk identification and compliance decision-making.
Automated KYC document extraction, Built a fraud detection model for credit card transactions
Built a predictive model for loan pay-ability using Logistic Regression, Random Forest, and XGBoost.
Improved report retrieval by 40% through an NLP-based chatbot integrated with TCS's advanced visualization tool, Vitellus.
Built a predictive model for loan pay-ability, Improved report retrieval through an NLP-based chatbot integrated with TCS's advanced visualization tool
The AI-first Recruitment Operating System that unifies and automates the entire hiring lifecycle for modern enterprises worldwide.
Dr Job FZ LLC. 2026 © All Rights Reserved
