Data Scientist (ML/Pandas/NumPy/IoT, Ukraine)

Tasks

• Analyze embeddings confusion matrices and model failure patterns
• Analyze high dimensional sensor and feature datasets
• Diagnose dataset shift and representation collapse
• Identify clusters anomalies and distribution gaps
• Implement data centric improvements into ML pipelines
• Improve weakly labeled data using clustering and pseudo labeling
• Investigate imbalanced noisy and mislabeled data
• Perform data mining on field data collections
• Perform model aware data analysis for XGBoost SVR and tree based models
• Support analysis for deep learning models including CNNs and Transformers
• Translate findings into recommendations for data filtering and relabeling

Perks/Benefits

• Christmas holidays
• Health insurance

Skills/Tech-stack

CNN | Clustering | DVC | Data Curation | Data Drift | Data Quality | Dimensionality Reduction | Domain Adaptation | Image analysis | Imbalanced Data | KNN | Label Noise | MLflow | Matplotlib | NumPy | PCA | Pandas | Pseudo-labeling | Python | Scikit-learn | Seaborn | Series analysis | Support Vector Regression | TSNE | Time Series | Time Series Analysis | Transformers | UMAP | Vector Regression | Video Analysis | Weights and Biases | XGBoost