Senior Data Scientist

Mô tả công việc

Responsibilities

1. Customer Analytics & CRM

• Build and deploy segmentation models (RFM/RFMT, CLV, engagement tiers) for omni-channel retention (dine-in, app, delivery).

• Design and maintain churn prediction, uplift modeling, and win-back triggers, tracking incremental lift.

2. Personalization & Recommendation

• Recommend voucher types, products, and timing via contextual bandits, hybrid recommenders, and rule-based systems.

• Ensure ROI and budget alignment for marketing incentives.

3. Forecasting & Operations

• Forecast traffic, sales, and ingredient demand at daily/weekly/monthly levels using ARIMA, Prophet, GBDT, or deep learning models.

• Optimize staff scheduling (shift assignment) under constraints such as skills, peak hours, and labor rules.

4. Experimentation & Causal Inference

• Design and analyze A/B tests, Difference-in-Differences, and causal ML models (DML, DR-Learner) to evaluate campaign or pricing impact.

5. MLOps & Delivery

• Deploy models to production (batch/API), manage feature stores, MLflow tracking, and monitor model drift/decay.

• Collaborate with Data Engineering and Product teams to ensure reliable data pipelines and mart/feature tables.

6. Communication & Business Partnering

• Translate business problems into ML approaches, articulate trade-offs between accuracy and speed, and communicate insights clearly to non-technical teams.

• Bachelor's degree in a quantitative field (e.g., statistics, economics, mathematics, business administration, marketing).

• 4+ years of experience as a Data Scientist in B2C, Retail, F&B, or E-commerce environments.

• Strong in Python (pandas, NumPy, scikit-learn; LightGBM/XGBoost preferred) and SQL (including window functions & optimization).

• Proven track record in:

• Machine Learning (classification/regression, feature engineering, model calibration).

• Forecasting (hierarchical, seasonal, event-based, backtesting).

• Experimentation (A/B test design, DiD, noise control).

• Data mining techniques to collect data, analyze and construct solutions for complex business problems.

• Working knowledge of Git, MLflow, Docker, and dashboard tools (Power BI, Metabase, or Looker).

• Product-oriented mindset with strong communication and documentation skills.

• Nice to have:

• Recommender systems (matrix factorization, two-tower, contextual bandit).

• Causal ML (EconML, DoWhy, DR-Learner).

• Optimization & Operations Research (OR-Tools, Pyomo).

• Big Data & Orchestration (PySpark, Airflow).

• API & Deployment (FastAPI, Kubernetes, streaming).