The company is focusing itself in building AI and Automation solutions for Retailers within the group.

As an R&D intern at DPV, you will be a key contributor in developing a cutting-edge inventory optimization system for one of Asia's largest retail chains, with over 25,000 stores. This is a highly focused role where you will work in the development of a predictive demand forecasting model and also build and tune the downstream constraint optimization engine that translates forecasts into actionable inventory decisions. This experience will provide valuable exposure to building successful, full-stack AI/ML solutions in a high-impact, real-world setting.Key Responsibilities

Under the guidance of our data scientists, you will be involved in the full process of developing and deploying a machine learning-powered system:

• Predictive Modeling: Develop and evaluate advanced machine learning regression models for demand forecasting.

• Optimization Engine Development: Design and implement the core constraint optimization logic, leveraging frameworks like Google OR-TOols or similar solvers to balance inventory costs, service levels, and operational constraints across 25,000+ stores.

• Data Pipeline: Perform large-scale dataset preparation, cleaning, and feature engineering on real-world retail data.

• Model Lifecycle: Conduct rigorous model training, evaluation, validation, and documentation for both the forecasting and optimization components.

• Gain exposure to the full-cycle considerations of an industrial AI/ML project, from experimentation to production readiness.

Candidate Requirements

• Proficiency in Python and strong experience with scientific computing packages (e.g., Pandas, Numpy, SciPy).

• Solid background in probability, statistics, and data analytics.

• Knowledge and experience in training and evaluating common ML regression models.

• Proficiency in Git version control system.

• Problem-solving aptitude, inquisitive mindset, and passion in data & AI/ML.

Highly Desirable

• Knowledge/experience in constraint optimization and optimization techniques (e.g., linear programming, mixed-integer programming).

• Familiarity with optimization solvers, particularly Google OR-Tools is highly preferred.

• Knowledge/experience in deep learning or clustering algorithms.

• Experience with high-performance data manipulation tools (e.g., Polars) or out-of-memory distributed computing (e.g., Dask).

• Experience with cloud platforms, particularly GCP.

• Experience with visualization tools (e.g., Matplotlib, Seaborn, Plotly).