Machine Learning Engineer

Work with raw data (including big data). Must have knowledge of preprocessing procedures and ability to prepare data for modeling. Experience in feature engineering. Develop and debug in Python, Java, C or C++. Proficient in git version control. R and Matlab are also relevant. Design and evaluate intrinsic and extrinsic metrics of your model's performance which are aligned with business goals. Define how we build and deploy systems around intelligent algorithms. Work with computational packages (TensorFlow, Theano, PyTorch, Keras, Scikit-Learn, NumPy, SciPy, Pandas, Pickle, Seaborn, statsmodels, NLTK). Use Natural Language Processing for Name Entity Recognition, Sentiment Analysis, and topic modeling. Familiarity with Latent Dirichlet Allocation. Assess opportunities and develop models to support Sanctions Screening alerting and disposition decisioning. Innovate solutions for various sanctions screening problems. Build model diagnostics, choice justification paradigms, testing and training approaches. Apply machine learning and statistical modeling methods for supervised and unsupervised learning. Tune and analyze algorithms and extract actionable insights. These methods include (but are not limited to) regression analysis, clustering, outlier detection, novelty detection, decision trees, collaborative filtering, nearest neighbors, support vector machines, ensemble methods and boosting, neural networks and deep learning, feature selection, and factorization methods. Effectively communicate and explain complex data products in simple terms.

#J-18808-Ljbffr