Senior Data Scientist

Monroe Consulting Group is recruiting on behalf of a #1 global market leader in digital shopper marketing. Our client is seeking a Senior Data Scientist based in Makati. The role offers a hybrid work setup (2-3 days onsite) with a night shift schedule.

Key Job Responsibilities:

• Data Quality & Pipeline: Audit and improve training data pipelines: identify problematic samples, fix category mislabeling, refine train/test stratification, and detect data leakage. Contribute to building robust, reproducible data workflows.
• Model Evaluation: Build and run reproducible evaluation workflows across regions, KPIs, and categories. Develop qualitative assessment frameworks that go beyond aggregate metrics to capture real-world prediction quality.
• Experimentation: Execute directed ML experiments: test new data splits, augmentation strategies, encoders, or hyperparameter configurations. Understand modern CV and multimodal techniques well enough to selectively adapt them when they solve a concrete modeling or product problem.
• Productionization: Take validated experiments and integrate them into the production training pipeline. Make existing research pipelines (e.g. saliency maps, visualization tools) repeatable and runnable by others. Ensure changes are tested, documented, and deployed without breaking existing functionality.
• Infrastructure: Support ML training infrastructure on Azure: launch training jobs, manage checkpoints, debug compute issues, and improve reproducibility of training runs.
• Ad Hoc Data Analysis: Answer analytical questions as they come up: pull data via SQL, investigate model behavior on specific segments, produce quick analyses and annotated tables for the consulting team. Be the person who can dig into the database and come back with a clear answer.

Key Job Qualifications:

• 3-5 years of experience in applied ML or ML engineering, with hands-on work in computer vision or multimodal models.
• Strong software engineering fundamentals: clean, modular, production-grade Python, not just notebooks. Experience working in existing codebases, writing well-structured packages, and shipping code through PRs.
• Fluent in the Python ML ecosystem (pandas/polars, PyArrow, scikit-learn, numpy) with strong SQL skills.
• Hands-on experience with deep learning frameworks (PyTorch preferred).
• Comfortable with ad hoc data analysis. Able to write SQL queries, pull and clean data, and produce quick quantitative answers to business questions.
• Experience with Azure ML or equivalent cloud ML platform (job orchestration, compute management, model registry).
• Proficiency with AI-assisted development tools (e.g., Claude Code, Cursor, Copilot) and LLM-driven workflows as part of everyday engineering practice.
• Strong written communication skills. Ability to document work clearly and communicate findings and blockers in writing.

Preferred Qualifications:

• Experience with transfer learning and fine-tuning vision encoders (CLIP, SigLIP, or similar) in multimodal pipelines.
• Experience with PyTorch Lightning, Optuna, MLflow, or similar training/experiment tracking frameworks.
• Experience with CI/CD pipelines (GitHub Actions) and containerized deployments.
• Familiarity with Databricks, Azure Blob Storage, or similar data platform tooling.
• Comfort reading ML research papers and selectively adapting techniques into existing systems. Publications or open-source contributions are a plus, but practical application matters more than academic output.
• Background in consumer research, behavioral science, or a domain where model predictions must be interpretable to non-technical stakeholders.