Headline

Data Science Manager helping businesses unlock insights from complex data & drive measurable impact

About

I build data science systems that businesses actually use — not just models that stay in notebooks. With 10+ years of experience (6+ in Data Science), I specialize in forecasting, supply chain analytics, and production-grade ML systems, delivering solutions that directly influence planning, revenue, and operations. Currently a Data Science Manager, I lead cross-functional teams of Data Scientists, Engineers, and Product teams to design and deploy scalable ML solutions — from demand forecasting to decision intelligence systems. Key highlights: • Led end-to-end supply chain planning solution replacing manual workflows, fully adopted by business users • Built multi-model forecasting systems (DeepAR, SARIMAX, XGBoost, Prophet) across SKU hierarchies • Delivered fraud detection system improving profits by 15%+ and reducing manual effort by 75% I work at the intersection of: → Business strategy → Machine learning systems → Scalable data platforms Currently exploring LLM/AI applications in real-world enterprise use cases, building PoCs to bridge the gap between GenAI hype and practical impact.

Location

Bengaluru

AI stack

claude code

claude

cursor

hugging face

langchain

langgraph

Work experience

Data Science Manager at MathCo, Jul 2024 - Present · 2 yrs 4 mos, Bengaluru, Karnataka, India, - Led end-to-end development of a supply chain planning platform for a global beverage company, replacing Excel-based workflows and achieving full business adoption - Led a cross-functional team of 10 (Data Scientists, Engineers, Product), delivering scalable ML solutions aligned with business planning cycles - Built multi-model forecasting system (DeepAR, ARIMA, statistical models) improving forecast accuracy across diverse SKU segments - Implemented a cluster-based SKU segmentation strategy to enable targeted forecasting, recognizing that a single model does not perform effectively across all SKUs. - Designed model selection logic based on data availability: 1. Deep learning models(DeepAR, DeepNPTS etc) for SKUs with 2+ years of historical data 2.Time-series models (ARIMA, Google Time Series Foundation Models) for SKUs with 1+ year of data 3.Statistical approaches (Holt winters exponential smoothing) for SKUs with limited history - Led the development of ETL pipelines that streamlined data ingestion from the client data lake, supporting downstream ML pipelines and application workflows in a production environment. - Delivered a scalable, production-ready solution that replaced manual, Excel-heavy weekly forecasting workflows and was fully adopted by business users as part of their regular planning process. - Acted as a primary point of contact for client and business stakeholders, translating planning requirements into data-driven solutions and ensuring alignment with business objectives.

Senior Associate at MathCo, Jun 2022 - Jul 2024 · 2 yrs 2 mos, Bengaluru, Karnataka, India, - Worked on a sales forecasting solution for a global CPG company, focused on the UK region, covering the top 5 subcategories across 6 major retailers. - Built forecasts at PPG–retailer–week level, enabling granular visibility and decision-making for business users. - Designed a hierarchical forecasting framework to balance accuracy and scalability: 1. Used Prophet models at the subcategory level to capture overall demand trends and seasonality. 2. Applied XGBoost at the PPG level to predict percentage contribution, incorporating rich feature sets to capture lower-level variations. - Combined top-down and bottom-up outputs to generate final PPG-level forecasts by applying predicted contributions to subcategory-level demand. - Developed what-if simulation capabilities, allowing users to apply different promotion scenarios at the PPG–retailer level and forecast future sales impact. - Collaborated closely with business stakeholders to ensure the solution aligned with commercial planning and promotion strategies.

Senior Consultant at EY, Jul 2021 - Jun 2022 · 1 yr, Bengaluru, Karnataka, India, • Collaborated with Fortune 500 companies to solve a wide range of data science and automation use cases across industries. • Designed and deployed an insurance fraud detection solution that reduced manual intervention by 75% and improved profit margins by 15%+. • Built and automated RPA workflows using Blue Prism for SAP and web-based applications, supporting automation initiatives in the banking and CPG sectors.

Consultant at EY, Nov 2019 - Jul 2021 · 1 yr 9 mos, Bangalore

Associate Consultant at EY, Nov 2018 - Nov 2019 · 1 yr 1 mo, Bengaluru, Karnataka, India

Software Engineer Analyst at Accenture, Jun 2017 - Nov 2018 · 1 yr 6 mos, Bengaluru, Karnataka, India, -Hands on experience in Automation Tools such as Blue Prism and UFT. -Automated Web, Windows and Oracle based applications. -Implemented End to End Robotic Process Automation (RPA) in a banking sector using Blue Prism. -Created Dynamic region mode process which was solution to dynamic objects in surface automation. -Broke down processes into reusable objects using Object studio and utilized them to build processes in Process Studio efficiently. -Created external scripts using C# and integrated with Blue Prism to achieve process automation.

Software Engineer at Accenture, Jan 2016 - Nov 2018 · 2 yrs 11 mos, Bengaluru, Karnataka, India, • Worked as RPA developer with expertise on Blue prism and UIpath automating applications such as SAP, BEx analyser, FileZilla, and Citrix. • Developed reusable objects using C# and VBA.

Education

Bachelor of Engineering - BE, Mechanical Engineering, Easwari Engineering College(SRM Group)

Bachelor of Engineering, Mechanics, Easwari Engineering College