Hareesh Kumar KB

Senior Oracle Retail Techno-Functional Consultant with 19 years of experience across ORMS, ReSA/OReSA, ORPM, and ORWMS, delivering global transformation programs for top retailers including Tesco, Supervalu, Abercrombie & Fitch, Ahold, Primark, Lifestyle, Neiman Marcus Group and ASOS. Expert in PL/SQL, Unix Shell, retail integrations, batch optimization, and solution design with a strong track record of automation, incident reduction, and system stabilization. Skilled in leading cross-functional teams, driving end-to-end solutioning, and aligning Oracle Retail processes with business goals. Currently pursuing PG program on AI and ML in Texas University of Austin in collaboration with Great Learning. Recently upskilled in AI/ML with hands-on exposure to predictive modelling, data analysis and machine-learning workflows, with a strong focus on applying AI-driven insights to demand forecasting, replenishment optimization and retail process transformation.

• Omni-Channel Strategy & Retail Transformation
• AI Tools Workshop - B-10X
• TCS AI Hackathon - RAG-based Data Quality Validator

• TCS Innovista Sparkle Award - Automation of Item Creation & Ranging for client NMG.
• Ahold Global CIO Recognition - Received ASM champ award for Reducing RMS incidents by 52%.
• Appreciation by Oracle Retail Delivery and Business Head for excellent contribution.

• SuperKart Sales Forecasting (Demand Prediction), Developed an end-to-end machine learning forecasting pipeline using engineered store, regional, and product-level features, improving quarterly revenue prediction accuracy and enabling data-driven inventory and procurement decisions., Benchmarked multiple regression models and optimized a Random Forest Regressor using RandomizedSearchCV, achieving strong performance (R² ≈ 0.93) and significantly reducing forecast error., Built a production-ready deployment stack with Streamlit UI, and Docker containers on Hugging Face Spaces, generating actionable insights for store expansion, pricing, and stock planning.
• GenAI Medical Assistant - RAG Model, Built a RAG-based AI solution for domain-specific document retrieval, demonstrating expertise in LLMs, vector search, document indexing, and workflow reasoning to address healthcare challenges to overcome information overload and provide healthcare professionals with fast, contextually accurate insights., Delivered a functional AI prototype demonstrating real-time medical knowledge retrieval and showcasing its potential to standardize care practices and improve patient outcomes.