Aniruddha Kumar

About Aniruddha Kumar

Data Scientist specializing in Demand Forecasting, Time Series Modeling (SARIMA, ARIMA, LSTM), and AI-powered Operational Intelligence for logistics and supply chain ecosystems.At FloData Analytics, I built scalable forecasting pipelines processing 1.3M+ package records across 10+ metros, achieving 90%+ prediction accuracy and delivering a 50% reduction in stem time and 36% reduction in driving distance — reshaping last-mile logistics operations for a hypergrowth delivery network.Key Expertise-Time Series Forecasting: SARIMA, ARIMA, LSTM, seasonality detection, anomaly detection-Geospatial & Operational AI: Hub clustering, route optimization, geospatial API development, ops cost optimization- MLOps & Data Engineering: GCP, BigQuery, Dataform, Secret Manager, Slack API integrations, scalable, production-ready data pipelinesOther impact projects- Developed a 90%+ accurate microbiome-based disease prediction system (Crohn’s, Diabetes, Obesity, Colorectal Cancer) for clinical deployment- Led water scarcity forecasting for Algeria & India (LSTM/ARIMA) predicting 89% water stress probability by 2040, integrating Google Earth Engine with AI dashboards.Value Proposition: I don’t just build ML models — I design full-stack AI systems that actively reduce logistics lead times, optimize operational KPIs, and integrate seamlessly with enterprise data workflows.Currently open to roles in Demand Forecasting, Operational AI, MLOps, and Geospatial Data Science within product-led, logistics tech, and AI-first companies.Let’s connect and discuss how AI can drive your operational performance.

Experience

1. Data Scientist - Operations Team

   Flodata Analytics

   Apr 2025 — Present · New Delhi, IN

   Problem: Demand volatility in 10+ American metros causing inconsistent package delivery times and high stem time in last-mile logistics. What I did (Methods/Tools): Developed demand forecasting models (Python, BigQuery) processing 1.3M+ package records, integrating real-time ops data streams and time series intelligence. Impact: Achieved 90%+ forecasting accuracy, enabling route and resource planning that cut stem time by 50% and reduced driving distance by 36%. Problem: Inefficient hub-to-hub distance optimization leading to high logistics cost and delays. What I did (Methods/Tools): Engineered nano-hub clustering algorithm using K-Means and Geospatial Analytics to optimize package routing. Impact: Reduced route overlap and improved load balancing, contributing to 20% reduction in delivery ops expenses. Problem: Absence of real-time demand cycle insights and anomaly alerts for logistics operations. What I did (Methods/Tools): Built Time Series Intelligence dashboards identifying seasonal trends and sudden demand spikes. Impact: Allowed operational teams to pre-empt resource allocation and shift strategies dynamically, improving operational uptime by 30%.

Education

• Kendriya Vidyalaya

  Higher Secondary, Physics, Chemistry and Mathematics

• Vellore Institute of Technology

  Computer Science Engineering Specialisation in Artificial Intelligence and Machine Learning, Computer Science