14 days ago

Data Scientist

Kuala Lumpur, MalaysiaMid · 2-5 yearsData Scientist

What you'll do

Improve models and algorithms to further optimize business outcomes.
Exploratory analysis: use data to suggest and prove hypotheses
Modeling: build optimization / predictive / statistical models to learn from data and estimate the unknowns - demand and sales forecasting, dynamic pricing for ancillary products, and demand planning
Data operations: query data, deploy models and automate pipelines in cloud
Set up sound time-based validation and honest baselines, and prove a model beats them before it ships.
Write clean, reviewable Python and SQL, merged through proper code review.
Help analyze live experiments and learn to spot a misleading readout.
Communicate findings clearly to technical and non-technical stakeholders.
Document work so a teammate can run and extend it without you.
Working with commercial teams to maximize the revenue by infusing AI & ML in their systems.

BS in Physics, Mathematics, DataScience or Engineering discipline
Up to 4 yrs relevant experience beyond first degree
Experience with common data science toolkits, programming languages (.py), visualisation tools and SQL/NoSQL databases.
Experience building production ML systems, beyond notebooks and Kaggle competitions.
Solid understanding of machine learning algorithms, XGBoost, LightGBM, neural networks, decision trees, with a clear grasp of why you tuned what you tuned.
Strong Python and hands-on experience with ML frameworks such as scikit-learn, TensorFlow, or PyTorch.
Demonstrable understanding of forecasting and regression pitfalls - lag feature leakage, target leakage in cross-validation, high-cardinality categorical handling, and the trade-offs between MAE, MAPE, and RMSE.
Ability to interpret models — SHAP, partial dependence, residual diagnostics — and explain results to non-technical stakeholders without dumbing them down.
Hands-on Google Cloud Platform experience, particularly BigQuery (window functions, partitioning, cost-aware SQL) and Vertex AI (training jobs, model registry, endpoints, pipelines).
Strong ability to implement, improve, and deploy ML and mathematical models in Python (Golang a plus for performance-critical services).
Experience productionizing models end-to-end, from SQL feature pipelines to deployed serving endpoints, on GCP using Vertex AI and BigQuery.
Conduct systems tests for security, performance, and availability of deployed models.

Experience with propensity / take-up (purchase-probability) models and probability calibration is a plus.
Exposure to time-series forecasting at scale (many related series), probabilistic forecasts, or demand that builds up toward a deadline is a plus.
Nice-to-have: deep learning for tabular and time-series problems (TFT, N-BEATS, NeuralProphet, TabPFN, Chronos); AutoML tooling such as PyCaret for rapid baselining.
Nice-to-have: experience with LLM-based or agentic tooling (LangGraph, MCP servers, prompt engineering for structured outputs, eval harnesses for LLM systems)

Summarised by NextRaise from the employer’s description, which follows in full below.

Job Description

Duties and Responsibilities

Improve models and algorithms to further optimize business outcomes.

Work across the following areas:

Modeling: build optimization / predictive / statistical models to learn from data and estimate the unknowns - demand and sales forecasting, dynamic pricing for ancillary products, and demand planning

Set up sound time-based validation and honest baselines, and prove a model beats them before it ships.

Working with commercial teams to maximize the revenue by infusing AI & ML in their systems.

Requirements and Qualifications:

BS in Physics, Mathematics, DataScience or Engineering discipline Up to 4 yrs relevant experience beyond first degree
Experience with common data science toolkits, programming languages (.py), visualisation tools and SQL/NoSQL databases.

Machine and Deep Learning :

Experience building production ML systems, beyond notebooks and Kaggle competitions.· Solid understanding of machine learning algorithms, XGBoost, LightGBM, neural networks, decision trees, with a clear grasp of why you tuned what you tuned.·
Strong Python and hands-on experience with ML frameworks such as scikit-learn, TensorFlow, or PyTorch.·
Demonstrable understanding of forecasting and regression pitfalls - lag feature leakage, target leakage in cross-validation, high-cardinality categorical handling, and the trade-offs between MAE, MAPE, and RMSE.·
Ability to interpret models — SHAP, partial dependence, residual diagnostics — and explain results to non-technical stakeholders without dumbing them down.·
Hands-on Google Cloud Platform experience, particularly BigQuery (window functions, partitioning, cost-aware SQL) and Vertex AI (training jobs, model registry, endpoints, pipelines).·
Experience with propensity / take-up (purchase-probability) models and probability calibration is a plus.·
Exposure to time-series forecasting at scale (many related series), probabilistic forecasts, or demand that builds up toward a deadline is a plus.·
Nice-to-have: deep learning for tabular and time-series problems (TFT, N-BEATS, NeuralProphet, TabPFN, Chronos); AutoML tooling such as PyCaret for rapid baselining.

Algorithm Engineering :

Strong ability to implement, improve, and deploy ML and mathematical models in Python (Golang a plus for performance-critical services).·
Experience productionizing models end-to-end, from SQL feature pipelines to deployed serving endpoints, on GCP using Vertex AI and BigQuery.·
Conduct systems tests for security, performance, and availability of deployed models.·
Develop and maintain design documentation, error analysis runbooks, and troubleshooting guides.·
Git-based workflows, CI/CD discipline, and code review hygiene.·
Monitoring discipline : drift detection, data quality checks, model performance tracking in production.·
Nice-to-have: experience with LLM-based or agentic tooling (LangGraph, MCP servers, prompt engineering for structured outputs, eval harnesses for LLM systems)

Airasia

Kuala Lumpur, Malaysia

Company facts come from this company's own listings. We only show what the postings themselves carry.