Summary :
We are seeking a highly skilled and experienced Senior Data Scientist to join our team. The ideal candidate will be a hands-on expert in developing and deploying deep learning-based forecasting models at scale. You will be responsible for the entire model lifecycle, from ideation and data preparation to model training, deployment, and monitoring in a production environment. This role requires a deep understanding of time series analysis, a passion for building innovative solutions, and the ability to work with large, complex datasets.
Responsibilities :
- Design, develop, and implement state-of-the-art deep learning forecasting models to solve critical business problems.
- Lead the entire model development lifecycle, including data cleaning, feature engineering, model selection, training, and validation.
- Build scalable and robust forecasting solutions capable of handling massive, high-dimensional time series data.
- Select and apply appropriate deep learning architectures for time series, such as LSTMs, GRUs, and Transformers.
- Utilize and evaluate various forecasting libraries and frameworks like Neural Forecast , GluonTS , TSAI , TSLib , Merlion , PyTorch Forecasting , Darts , Orbit and others.
- Collaborate with data engineers to build and maintain the necessary data pipelines and infrastructure for large-scale model training and deployment.
- Perform rigorous model evaluation using appropriate time series-specific techniques like walk-forward validation and backtesting.
- Communicate complex technical concepts and model performance to both technical and non-technical stakeholders.
- Stay up-to-date with the latest research and advancements in deep learning for time series forecasting.
Required Qualifications :
Hands-On Experience ON LARGE DATASETS is a Must : Proven track record of building and deploying deep learning models in a production environment. Please be prepared to discuss specific projects and your role in them.Deep Learning for Time Series : Extensive experience with deep learning architectures tailored for time series data (e.g., LSTMs, GRUs, and especially Transformer-based models like Temporal Fusion Transformer, Autoformer, or PatchTST).Programming Proficiency : Expert-level proficiency in Python and its data science ecosystem (Pandas, NumPy, Scikit-learn).Forecasting Libraries : Hands-on experience with at least two of the following libraries :TSLibPyTorch ForecastingDartsNeuralforecastGluonTSTSAIMerlionFBProphetPytorch ForecastingOrbitFrameworks : Strong experience with deep learning frameworks such as PyTorch or TensorFlow.Scalability : Demonstrated ability to build and train models at scale, including experience with global models and distributed computing.Statistical and Analytical Skills : Solid understanding of time series statistics, including seasonality, trends, and autocorrelation.Collaboration & Communication : Excellent communication skills with the ability to work effectively in a cross-functional team.Preferred Qualifications :
Advanced degree (M.S. or Ph.D.) in a quantitative field such as Computer Science, Statistics, Mathematics, or a related discipline.3 years experience in the forecasting domain.