Anupam Srivastav

Experienced Data Scientist with a strong background in Data Analytics, Machine Learning, and Image Analytics. Proven ability to deliver actionable insights, optimize processes, and drive business growth. Adept at leading cross-functional teams and managing end-to-end data science projects.

Silicon Prediction Model for Blast Furnace

- Developed a model to predict the amount of silicon in hot metal 2 hours in advance, providing control levers for temperature control and coke rate consumption.

Heating Model for Coke Ovens

- Created a model to predict actual battery temperature using regenerator temperature and suggest fuel consumption based on it, optimizing fuel use and maintaining desired temperature levels.

Raw Material Handling System

- Optimized scheduling for the coal circuit to ensure a consistent supply while maintaining blend ratio and equipment utilization.

- Improved wagon tippler turnaround time to empty wagons within specified limits, preventing demurrage charges and ensuring operational efficiency.

Machine Learning Based Secondary Charge Calculation

- Created an ML model to predict Iron Ore requirements, enabling data-driven decisions for Primary Steel Making. Resulted in a 10% reduction in material wastage.

Quality Prediction Model for New By-Product Plant

- Developed a regression model to predict Lean Liquor and Stripped Water quality, leading to a 15% improvement in temperature control accuracy before lab analysis.

Anomaly Detection in Process Equipment Causing Stack Emission

- Implemented an anomaly detection system for stack emissions, improving predictive maintenance and reducing emission deviations by 20%.

Charging Emission Detection in Coke Plant

- Built a system for fugitive emission detection during oven charging, facilitating timely maintenance and reducing environmental impact by 18%.

Operational Delay Insight Generation

- Applied text summarization techniques to identify top delay reasons, enabling better maintenance planning and reducing downtime by 12%.

Code Genie

- Developed a coding assistant using multiple Large Language Models (LLMs) capable of helping with code, debugging, and interacting with attachments.

- Significantly reduced IT costs by automating coding assistance, improving code quality, and reducing debugging time.

-IBM Data Science Professional Certificate (https://coursera.org/verify/professional-cert/LAN7SWDL5CPP)

IBM AI Engineering Professional Certificate

(https://www.credly.com/badges/f8a7c3f6-c5d2-4dbe-8928-2dfe19b3f5f0/public_url)