About
Highly accomplished Data Science leader with nearly a decade of experience in driving product innovation across diverse sectors including CPG, BFSI, and IET. Proven expertise in leveraging AI/ML, Generative AI, and NLP to build scalable, high-impact products and customer solutions. Adept at all stages of product development, from ideation to deployment, with a strong command of Python, SQL, and advanced machine learning techniques.
Work
Gurugram, Haryana, India
→
Summary
Guided CXOs on AI/ML initiatives and owned the vertical for solutioning and deployment, driving product innovation for the travel sector.
Highlights
Guided the development of a Neo-Generative AI Chatbot for Booking Agents, integrating structured and unstructured data to reduce customer queries and improve agent stickiness.
Delivered a Lead Generation & Scoring Engine that increased Monthly Onboarded Users (MOUs) by 60% within the launch month, significantly boosting user acquisition and T1 bookings.
Led fine-tuning of Qwen 3 LLM using proprietary data to align responses with TBO's business context, improving accuracy and relevance of agent-assist, eventually consumed within the Neo App.
Gurugram, Haryana, India
→
Summary
Led Generative AI solution development and MLOps platform implementation, driving efficiency and insights for major CPG, retail, and flight companies.
Highlights
Led a team to develop and deploy a Generative AI-powered Call Centre Agent Assist solution on Azure, reducing Average Handle Time (AHT) by 20% and agent onboarding time by 40%.
Delivered Envision Bot, a chatbot leveraging a multi-agent framework & LLMs deployed on Azure, to retrieve insights from PowerBI, SQL, and documents, reducing time to insights by 85%.
Designed and implemented a large-scale calendar optimization solution on AWS for 20 retailers and 2000 SKUs, generating 52-week promo calendars and driving a 23% uplift in ROI.
Directed a cross-functional team (11 members) to architect and implement an in-house Model Observability Platform, onboarding 18 production ML models and reducing model downtime and issue triage time by >40%.
Gurugram, Haryana, India
→
Summary
Designed and implemented advanced AI/ML solutions for CPG manufacturers, focusing on marketing content creation and time-series forecasting.
Highlights
Designed and implemented a personalized Marketing Content Creation tool using LLMs and prompt engineering, reducing content creation time by 90% for a major CPG company.
Developed and implemented a scalable time-series forecasting solution alongside media attribution, price, and promo optimization, resulting in ~15% more accurate demand predictions and a ~26% increase in ROI.
Gurugram, Haryana, India
→
Summary
Led Data Science teams under the Financial Planning and Analytics domain, responsible for planning, execution, and delivery of client solutions.
Highlights
Provided Data Science content and strategic guidance to Account Leads and Stakeholders, facilitating RFP responses and client problem-solving.
Developed an Insurance Forecasting platform, an accelerator used to deliver on client needs by Single Deal Value.
Implemented ServiceNow Auto ticket classification, categorizing incoming tickets into pre-defined Configuration Items and Symptom Codes for a major CPG client.
Engineered an Employee Turnover Prediction model, accurately predicting employee churn to enable proactive retention strategies.
Gurugram, Haryana, India
→
Summary
Led the Data Science development of Foresient, an industry-agnostic automated time-series forecasting platform deployable across multiple cloud environments.
Highlights
Led the end-to-end Data Science development of Foresient, an industry-agnostic automated time-series forecasting platform, deployable across AWS, GCP, Azure, and on-Prem.
Gurugram, Haryana, India
→
Summary
Developed advanced machine learning models for customer matching and predictive analytics, enhancing insights and prediction transparency.
Highlights
Developed a synthetic matching algorithm for customer matching, improving customer insights and engagement.
Created an uplift machine learning model for end-to-end training and hyper-parameter optimization, utilizing interpretable ML techniques to enhance prediction transparency.
Gurugram, Haryana, India
→
Summary
Built and deployed personalized recommendation, click-through rate, and churn prediction engines for enhanced user engagement and retention.
Highlights
Built and deployed a personalized job recommendation engine using collaborative filtering and deep learning techniques.
Designed and implemented advanced machine learning models to predict click-through rates, improving campaign targeting and achieving a ~12% increase in ad engagement.
Developed and deployed a machine learning model to identify customers at risk of churn, enabling targeted retention strategies and reducing churn rates by ~10%.
Skills
Programming Languages & Tools
Python, R, SQL, Pytorch, PySpark.
Machine Learning
Supervised Learning (Regression, Classification), Unsupervised Learning (Clustering, Dimensionality Reduction), Deep Learning (CNN, RNN, GAN, Transformers), Optimization (LP, NLP, MILP, Bayesian, Genetic Algorithm).
Technologies & Platforms
Azure, AWS, PowerBI, LLMs, NLP, Generative AI, RAG, MLOps, API, Telegram, Reddit, GitHub, LinkedIn, Medium.