Rohit Agarwal

• Developed & deployed scalable forecasting pipeline using ESM, LGBM & Sequence to Sequence Neural network architecture like LSTM, GRU with embeddings.
• Established framework to address stability, accuracy & anomaly detection in demand forecasting.
• End to End implementation of Demand forecasting module using Argo for millions of Time series combinations.

Tools & languages used- Pyspark, SQL, Argo, Python

• Common stock pooling - An optimization solution & report that provides the estimated availability of common stock to any territory basis historical selling capacity.
• Workforce Management - Created a staff scheduling tool & KPI tracker that provides staff schedule basis hourly forecasted footfall/ invoices at store level. Received a flash award for implementing this project in CenterPoint stores of Landmark group Dubai. (Model used - SVR, MLR, NL optimization)
• Drafted daily and monthly reports on supply chain KPI's to assess successes and anomalies.

Tools & languages used - Python, R, SQL, SAS

• Developed model to estimate the no of days required for the delivery of couriered medicines. (LR model)
• Developed & deployed online payment reconciliation & anomaly detection automated triggers.
• Developed an optimization framework for determining per order cost to be paid to delivery boys. Resulted in reduction of logistics costs by 15%. (Linear optimization)
• Synthesized current business intelligence data to produce reports and polished presentations, highlighting findings & recommending changes.

Tools & language used - Python, R, SQL, Spotfire

• Created reports of patients who are at high risk of readmission in hospitals & identified important business KPI's for the same.
• Created various Excel documents to assist with pulling metrics data and presenting information to stakeholders for concise explanations of fluctuations in KPI's
• Developed tables, views and materialized views using SQL.

Tools & Languages - R, SQL & Spotfire