Search by job, company or skills

neoqubit

Senior Data Scientist

Save
  • Posted 23 hours ago
  • Be among the first 10 applicants
Early Applicant

Job Description

We are seeking a PhD-trained statistician to develop, validate, and deploy demand forecasting and related predictive models for complex, large-scale systems. The role focuses on modeling demand under varying constraints, understanding substitution and recapture behavior, and supporting downstream planning and optimization decisions through robust statistical and probabilistic models. This position is ideal for a statistics PhD who enjoys applying rigorous statistical methods to real-world forecasting, decision support, and large-scale data problems.

The Core Responsibilities For The Job Include The Following

Demand Forecasting and Predictive Modeling:

  • Develop and maintain demand forecasting models at multiple levels of aggregation (e. g., temporal, product, market, or network level).
  • Model unconstrained vs. constrained demand, accounting for capacity limits, spillover effects, and substitution and recapture behavior.
  • Produce short- and medium-term forecasts over multiperiod horizons (e. g., daily to weekly forecasts).

Statistical Modeling And Inference

  • Apply a range of statistical techniques, including time-series models (e. g., ARIMA, state-space, and hierarchical time series); regression and generalized linear models; Bayesian models; and probabilistic forecasting.
  • Quantify forecast uncertainty and communicate confidence intervals and risk measures clearly to stakeholders.

Machine Learning For Forecasting

  • Develop and evaluate ML-based forecasting models, such as tree-based models, regularized regression, and neural networks (where appropriate).
  • Compare ML and classical statistical approaches, selecting models based on: Accuracy, stability, and interpretability.
  • Support hybrid approaches where ML outputs feed into downstream analytical or optimization models.

Model Validation And Performance Monitoring

  • Design rigorous model validation frameworks, including backtesting, cross-validation, and error decomposition.
  • Monitor model performance over time and diagnose drift or breakdowns in predictive accuracy.
  • Recommend model recalibration or redesign when underlying demand patterns change.

Collaboration With Optimization And Planning Teams

  • Work closely with optimization, planning, and analytics teams to: Use supply and demand forecasts as inputs to decision models.
  • Provide statistical insights on model assumptions and limitations.
  • Support integrated workflows where forecasting informs capacity allocation and scenario analysis.

Data Analysis And Insight Generation

  • Conduct exploratory data analysis on large datasets to identify demand patterns, structural changes, seasonality, and anomalies.
  • Translate statistical findings into actionable business insights.

Requirements

  • Strong statistical foundations with applied modeling experience.
  • Comfortable working with imperfect real-world data.
  • Thoughtful about uncertainty, assumptions, and model limitations.
  • Able to explain complex statistical ideas in simple, practical terms.
  • Interested in influencing decisions through high-quality forecasts.
  • Skills Required: Time Series Forecasting, Bayesian Statistics / Probabilistic Modeling, Python (statsmodels, scikit-learn, and PyMC), Demand Forecasting, Statistical Modeling, ARIMA / State Space Models, ML Model Validation, and Regression Modeling.

Education And Statistical Foundations

  • PhD in Statistics, Biostatistics, Applied Mathematics, Econometrics, or a closely related field.
  • Strong foundation in probability theory, statistical inference, regression modeling, and time-series analysis.

Forecasting And Modeling Experience

  • Hands-on experience building forecasting models for real-world datasets.
  • Familiarity with constrained vs. unconstrained demand modeling and substitution or recapture effects (broadly defined).
  • Ability to balance model accuracy, robustness, and interpretability.

Programming And Tools

  • Strong programming skills in Python.
  • Experience with relevant libraries (e. g., statsmodels, scikit-learn, PyMC, Stan, or equivalent).
  • Comfortable working with large, messy datasets.

Preferred / Nice To Have Qualifications

  • Experience working on forecasting or analytics problems in applied domains such as planning, resource allocation, and supply-demand systems.
  • Familiarity with optimization concepts and how forecasts are used in decision models.
  • Experience communicating statistical results to nontechnical audiences.

What Makes This Role a Good Fit for a Statistics PhD:

  • Strong emphasis on core statistical thinking and inference.
  • Opportunity to apply theory to real, impactful forecasting problems.
  • Balance between classical statistics and modern ML methods.
  • Clear pathway to influence downstream decision-making without owning optimization logic.
  • Structured, well-scoped problems with room for methodological depth.

This job was posted by Arulsakthi V from Neoqubit.

More Info

Job Type:
Industry:
Function:
Employment Type:

About Company

Job ID: 148888893

Similar Jobs

Bengaluru, India

Skills:

CassandraPysparkSvmTableauTensorflowNumpyPythonHadoopHBaseSqlArtificial Neural NetworksNaive BayesHivePandasXGBoostSparkDatabricksKerasMongoDBAws S3machine learning techniquesk-NNDecision ForestsMap ReduceNoSQL databasesScikit-learnSageMakerAthena

Bengaluru, India

Skills:

causal inference SqlPythonMachine LearningOptimisation Techniques

Bengaluru, India

Skills:

data wrangling TensorflowPytorchData VisualizationPythonMachine Learning AlgorithmsAWS BedrockGCP Vertex AIscikit-learnstatistical methodsmachine learning models

Bengaluru, India

Skills:

snowflake CassandraKafkaTensorflowNumpyPytorchDockerFlaskPythonMatplotlibBigQueryJiraRedisSqlJenkinsGitPandasSparkDatabricksApache BeamKerasKubernetesFlinkPub SubKubeflowTFXJupyter

Bengaluru, India

Skills:

PytorchSqlPythonTensorflowData Analytics Techniques