Senior Data Scientist, Deep Learning Forecasting

About CAI Stack CAI Stack delivers modular AI infrastructure that enables enterprises to build, scale, and deploy advanced ML and DL solutions. Our platform powers vertical-specific AI applications for some of the world's leading organizations, helping them achieve faster, more accurate business decisions. Role Overview We are looking for a Senior Data Scientist to drive deep learning-based forecasting initiatives. This role involves designing, training, and deploying predictive models capable of handling complex, high-dimensional time series data. You will oversee the entire lifecycle of forecasting models, ensuring robustness, scalability, and actionable insights for business stakeholders. Key Responsibilities Develop state-of-the-art deep learning models for business forecasting. Lead end-to-end model development including data preprocessing, feature engineering, architecture selection, training, and evaluation. Design scalable solutions for large, multi-dimensional time series datasets. Select and implement advanced architectures such as LSTMs, GRUs, and Transformer-based models. Work with libraries and frameworks like Neural Forecast, GluonTS, TSAI, TSLib, Merlion, FBProphet, PyTorch Forecasting, Darts, Orbit, and statsmodels. Collaborate with data engineers to build pipelines and infrastructure for efficient model training and deployment. Apply rigorous evaluation methods including walk-forward validation and backtesting. Effectively communicate model insights and results to technical and non-technical stakeholders. Stay current with research and advancements in deep learning for time series forecasting. Required Qualifications Proven experience in building and deploying deep learning forecasting models in production. Strong expertise in time series architectures such as LSTMs, GRUs, and Transformer-based models (e.g., Temporal Fusion Transformer, Autoformer, PatchTST). Advanced proficiency in Python and relevant data science libraries (Pandas, NumPy, Scikit-learn). Hands-on experience with at least two of these forecasting libraries: TSLib PyTorch Forecasting Darts Neuralforecast GluonTS TSAI Merlion FBProphet Orbit Experience with recommendation or specialized ML libraries (at least two): TorchRec TensorFlow Recommenders (TFRS) NVIDIA Merlin Microsoft Recommenders NVIDIA Recommenders Alibaba Recommenders Expertise in deep learning frameworks such as PyTorch or TensorFlow. Experience scaling models using distributed computing and global model strategies. Strong understanding of statistical principles for time series data including trends, seasonality, and autocorrelation. Excellent communication and teamwork skills for cross-functional collaboration. Preferred Qualifications Advanced degree (M.S. or Ph.D.) in Computer Science, Statistics, Mathematics, or related quantitative field. At least 3 years of experience in the forecasting domain.