AI / ML Data Scientist - Time Series

Build forecasting solutions on large datasets and combine predictive models with generative AI reasoning to produce business insights, using time-series forecasting, distributed data processing, statistical modeling, and generative AI.

Responsibilities:

• Build scalable data-processing pipelines with PySpark on Google Cloud Dataproc.
• Perform feature engineering, time-window aggregation, and KPI calculation in PySpark.
• Conduct feature selection and feature-importance analysis.
• Develop and optimize time-series forecasting models using ARIMA, SARIMA, SARIMAX, and XGBoost.
• Work with BigQuery and SQL-based datasets.
• Integrate Gemini into analytics and AI workflows.
• Develop LLM applications with LangChain and apply prompt engineering to produce contextual business insights.
• Combine forecasting outputs with generative AI reasoning.
• Collaborate with business and engineering teams to move AI solutions into production.

Requirements:

• 3-7 years of overall professional experience, including 1-3 years of directly relevant hands-on experience.
• Strong Python, PySpark, and PySpark ML skills, including experience with Google Cloud Dataproc.
• Experience with scikit-learn, statistical modeling, and time-series forecasting.
• Hands-on knowledge of ARIMA, SARIMA, SARIMAX, and XGBoost.
• Experience with BigQuery, SQL, and GCP data and AI services.
• Understanding of LLMs, LangChain, and prompt engineering.

Nice To Have:

• Experience building production ML pipelines.
• Experience with model monitoring and evaluation.
• Experience combining generative AI with predictive analytics.
• Experience working with large-scale enterprise datasets.