About the job
- Job Title: Contract Data Scientist | ML Engineer (Time Series Forecasting) - 3 Months
- Location: Kathmandu, Nepal
- Job Type: Contract (3 Months)
- Pay Range: NPR 20,000 - 35,000 per month
About the Role
We are seeking a skilled Data ML Engineer with expertise in time series forecasting and ensemble model development to join our team on a 3-month contract. The ideal candidate will have a strong statistical background, hands-on experience deploying machine learning models in Snowflake, and a passion for contributing to open-source communities. This role requires timely delivery of production-ready code and a deep understanding of end-to-end model training and deployment workflows.
Key Responsibilities
- Design, develop, and deploy ensemble models in Snowflake for time series forecasting tasks.
- Collaborate with cross-functional teams to integrate ML solutions into production pipelines.
- Contribute daily to open-source forums (e.g., GitHub, Kaggle, Stack Overflow) to stay updated and share insights.
- Ensure code is optimized, scalable, and delivered on schedule for deployment.
- Stay updated with emerging technologies and best practices in time series forecasting.
- Document workflows, model performance, and deployment processes.
Requirements
Must-Have Skills:
- Proven experience with time series forecasting (e.g., ARIMA, Prophet, LSTM).
- Expertise in building and deploying ensemble models (bagging, boosting, stacking).
- Strong statistical background (hypothesis testing, distributions, regression).
- Hands-on experience with Snowflake for data warehousing and ML deployment.
- Proficiency in Python/R and ML libraries (Scikit-learn, TensorFlow, PyTorch).
- Familiarity with MLOps tools (MLflow, Kubeflow) and version control (Git).
- Active participation in open-source communities (share portfolio/GitHub profile).
- Ability to write clean, production-ready code under tight deadlines.
Preferred Skills:
- Experience with cloud platforms (AWS, Azure, GCP).
- Knowledge of Snowpark, Snowflake’s ML features, or dbt.
- Familiarity with AutoML tools (H2O, TPOT) for rapid prototyping.
Important Time Series Forecasting Technologies
Candidates should be familiar with:
- Classical Methods: ARIMA, SARIMA, Exponential Smoothing.
- Machine Learning: Prophet, XGBoost, LightGBM, CatBoost.
- Deep Learning: LSTM, GRU, Transformer-based models.
- Tools: TensorFlow, PyTorch, Darts, Statsmodels.
- Cloud/MLOps: Snowflake ML, AWS SageMaker, MLflow, Airflow.
Why Join Us?
- Work on cutting-edge ML projects with real-world impact.
- Flexible remote/hybrid work options.
- Opportunity to grow your open-source portfolio.