Sunayana Jindal

I’m a Senior ML Engineer with 3+ years of experience building scalable forecasting solutions at Samsung Research. I focus on translating data signals into production-ready models that improve outcomes across real business constraints.

At Samsung R&D Institute India (SRIB), I contributed as a core engineer to forecasting work recognized as Samsung’s “New Valley Project of the Year, 2025”. I’ve led efforts to optimize ad campaign overdelivery, saving $18M in revenue every half-year, and improved sparse-data forecasting in low-traffic markets by reducing prediction error from 41% to 7.6%.

I take ownership across the full ML lifecycle—data processing, feature engineering, model training, hyperparameter tuning, backtesting, and validation—to ensure fast retraining cycles and strong model performance. I also build the systems that make models operational, including MLOps infrastructure for automated end-to-end workflows and CI/CD to support continuous learning.

I enjoy combining deep learning with practical engineering for high-impact results, whether it’s time-series transformers for forecasting at scale (e.g., 750M impressions) or hierarchical neural architectures that reduce forecast error across 38 countries during peak periods. I’m especially motivated by cross-team collaboration and research-driven improvements, including co-authoring a hybrid ML approach paper under review.