Tech Stack
Machine LearningPython
Job Description, Responsibilities & Requirements
About the Position
We are looking for a Senior Data Scientist to join our fintech team, focusing on pricing optimization for insurance companies using an ensemble of predictive models. The tech stack includes: pandas, numpy, PyTorch, SHAP, and scikit-learn.
Responsibilities
- Create statistical summaries to support hypothesis testing and data-driven decisions during EDA
- Implement data preparation and feature engineering pipelines for the models
- Plan and implement algorithms for predictive modeling
- Provide continuous improvement of models in production
- Lead the ML-driven Python coding for end-to-end model deliveries
- Support our ML-related libraries
- Collaborate with a cross-functional team of specialists, including senior and junior data scientists, Python developers, ML engineers, data analysts, and DevOps specialists
- Provide expertise in tuning loss functions, metrics, sample weights adjustment, hyperparameter tuning, and model improvement in general
- Generate detailed evaluation reports, summarizing key findings and presenting actionable insights to stakeholders
- Keep up with tight deadlines, agile environment of work with evolving objectives and KPIs, having the highest level of organization and self-management to provide full work clarity, extensive tracking, and documentation of your work
Requirements
- Experience with Time Series: Deep understanding of time series forecasting (Trend, Seasonality, Stationarity). Experience with ARIMA/SARIMA, Prophet, Exponential Smoothing models.
- Python Stack: Proficiency in Pandas, NumPy, Scikit-learn. Experience with deep learning frameworks (PyTorch or TensorFlow) for building recurrent networks (LSTM, GRU).
- Scoring & Classification: Ability to build scoring models, work with logistic regression, decision trees, and ensembles (XGBoost, CatBoost, LightGBM).
- Data Engineering: Skills in SQL (PostgreSQL/MySQL), handling large volumes of data, cleaning from "noise" and filling in missing values.
- MLOps: Understanding of the model lifecycle - from experiments in Jupyter to deployment in Production (Docker, FastAPI).
We Offer
- Competitive remuneration package
- Bonuses
- Professional mentorship and guidance from experienced team members
- Opportunities for professional growth and continuous learning
- A dynamic and collaborative work environment
About the Company
DataObrii is a high-tech consulting firm specializing in data science, machine learning, and AI-augmented Internet of Things. Our team comprises experienced data scientists, Python engineers, DevOps, hardware electrical engineers, and business analysts dedicated to delivering innovative, data-driven solutions that enhance business intelligence and efficiency.
Our Values:
- Efficiency: We employ an agile approach to ensure timely delivery of high-quality solutions.
- Professionalism: Our commitment to excellence drives us to achieve success for our clients.
- Creativity: We utilize design thinking to develop innovative solutions that address complex business challenges.
- Care: We invest time in understanding our clients' business models to provide tailored solutions that align with their objectives.