AI/ ML Engineering
Nexxoft Infotel Limited · Sep 2025 – Aug 2026
- Developed an end-to-end real-time sales forecasting solution to predict future sales using historical transaction data, seasonal trends, promotions, and customer demand patterns.
- The project automated data ingestion, preprocessing, model training, prediction, and dashboard reporting to help business teams optimize inventory, improve demand planning, and increase revenue.
- Gathered sales data from SQL databases and AWS S3 for analysis and model development.
- Performed exploratory data analysis (EDA) to identify sales trends, seasonality, and customer buying patterns.
- Developed feature engineering pipelines using historical sales, promotions, holidays, and product categories.
- Built and trained machine learning models for sales forecasting using Scikit-learn and XG Boost.
- Evaluated model performance using MAE, RMSE, and MAPE metrics, and optimized model accuracy through hyperparameter tuning.
- Automated data preprocessing and prediction workflows using Python scripts and AWS Lambda.
- Deployed the forecasting model on AWS SageMaker/EC2 for real-time predictions.
- Wrote optimized SQL queries to extract, aggregate, and analyze sales data.
- Stored datasets, trained models, and prediction outputs in AWS S3.
- Monitored model performance using ML flow and AWS CloudWatch, and retrained models when required.
- Created interactive dashboards in Power BI/Tableau to compare forecasted and actual sales.
- Collaborated with business stakeholders to understand forecasting requirements and deliver actionable insights.
- Used Git for version control and followed Agile/Scrum methodologies for project development.
- SQL & data querying: Write queries to extract and validate datasets assist in building datasets for analysis.


