Requirements: English
Company: SquareOne
Project Overview:
You will join theTime Series Forecasting Toolbox (TSFT) Team, part of a strategic initiative to build a reusable, scalable Python library designed to support business-critical time series forecasting. This project is embedded within a broader ecosystem of AI Analytics platforms focusing on observability, knowledge access, intelligent automation, and data governance.
TSFT combines traditional statistical forecasting, machine learning, and deep learning approaches (e.g., scikit-learn, statsmodels, statsforecast, neuralforecast, PyTorch). The roadmap includes incorporating LLM-based forecasting capabilities. The work involves close collaboration with cross-functional teams, contributing to platform components that support a wide range of enterprise applications.
You will also be involved in other strategic areas such as:
- Grafana + OpenTelemetryfor full observability and user behavior tracking
- Sinequa Enterprise Searchfor knowledge centralization
- AI/ML tooling(e.g., transformers, scikit-learn) for behavior modeling and segmentation
- AI agents(LangChain, CrewAI) for intelligent automation
- Data Governance FAIR principles(DataHub, Collibra)
- Data storytelling and visualization (Grafana, Streamlit, Power BI)
Must Have:
- 3+ years of professional experience in Python-based machine learning projects
- Solid understanding of machine learning algorithms (supervised learning, hybrid models, overfitting, etc.)
- Proven ability to write clean, efficient, object-oriented Python code
- Practical experience with model evaluation and testing techniques
- Basic understanding of time series analysis principles
Should Have:
- Experience with structuring ML solutions into scalable software packages
- Exposure to or interest in time series forecasting libraries/methods
- Familiarity with deep learning libraries relevant to forecasting (e.g., PyTorch, neuralforecast)
- Understanding of CI/CD pipelines, unit testing, and MLOps principles
Nice to Have:
- Experience with containerization tools (e.g., Docker)
- Exposure to enterprise-grade platforms (Grafana, Streamlit, Power BI, LangChain, Collibra)
- Knowledge of FAIR data principles and data governance practices
Project Overview:
You will join theTime Series Forecasting Toolbox (TSFT) Team, part of a strategic initiative to build a reusable, scalable Python library designed to support business-critical time series forecasting. This project is embedded within a broader ecosystem of AI Analytics platforms focusing on observability, knowledge access, intelligent automation, and data governance.
TSFT combines traditional statistical forecasting, machine learning, and deep learning approaches (e.g., scikit-learn, statsmodels, statsforecast, neuralforecast, PyTorch). The roadmap includes incorporating LLM-based forecasting capabilities. The work involves close collaboration with cross-functional teams, contributing to platform components that support a wide range of enterprise applications.
You will also be involved in other strategic areas such as:
- Grafana + OpenTelemetryfor full observability and user behavior tracking
- Sinequa Enterprise Searchfor knowledge centralization
- AI/ML tooling(e.g., transformers, scikit-learn) for behavior modeling and segmentation
- AI agents(LangChain, CrewAI) for intelligent automation
- Data Governance FAIR principles(DataHub, Collibra)
- Data storytelling and visualization (Grafana, Streamlit, Power BI)
,[Develop and maintain reusable Python-based forecasting components using object-oriented programming, Work across the full machine learning lifecycle: from data preparation and modeling to productization and operationalization, Apply best practices in code design, testing (unit/integration), and DevOps/CI/CD workflows, Support implementation of new methodologies including LLMs for forecasting, Collaborate with platform and data engineering teams to ensure scalable deployment, Contribute to internal knowledge sharing and documentation] Requirements: Machine learning, Python, Testing, Forecasting, Deep learning, PyTorch, CD pipelines, MLOps, Docker, Grafana, BI