Data Scientist / Decision Scientist

Key Skills

Python Programming, SQL, Cloud computing, Apache Hadoop, Machine learning techniques, Digital Innovation, JIRA, Bayesian networks, Data Visualization, Deep Learning, ETL/ELT technologies, Data Modelling, SPARK Programming, Neural Networks, Data Lake, Time series, Linear Models, Gradient Boosting trees, Decision trees, Open source project tools, Decision Science

Job Description

Data Scientist

Responsibilities:

• Architect a data based solution for the business problem presented. Develop contextual automotive domain knowledge around the problem. Collaborate with the domain knowledge expert if required.
• Understand the IoT device (Telematics Control Unit) of the vehicle. Develop understanding of Time-Series Data originating from TCU.
• Understand the data landscape, establish the data adequacy beforehand
• Clean up and Prepare datasets for modeling, get involved in ETL process if required. Apply data transformation techniques such as resampling, filtering, encoding etc. 
• Exploratory data and get insights. Present the descriptive stats and insights to the domain experts. Find meaningful patterns in data, detect seasonality and trend, establish cause and effect relationships in data. Develop & Test hypothesis in collaboration with the domain experts.
• Design features, shortlist features, study feature importance, decide the ML strategy
• Data modelling, selection of an appropriate machine learning / deep learning model, data pipeline setup for model training, hyper-parameter tuning, validation and test. Apply ensemble model techniques (if required) 
• Reporting & Visualization: Comprehension of reports, visualization of data in the form of plots. Creating Heat Maps using Google Maps API
• Mentoring juniors in data

Technical Skills / Experience

Essential:

• Must have minimum of 3 years of industry experience in developing data science models.
• Experience using machine learning algorithms (e.g. Generalized Linear Models, Boosting, Decision Trees, Neural Networks, SVM, Bayesian Methods, time series models, etc.)
• Hands-on experience in using machine learning models for regression and classification problems. 
• Hands on experience in unsupervised machine learning algorithms. Working knowledge of clustering techniques such has, centroid based clustering.
• Working experience with Cloud Computing Platforms such as AWS / IBM Strong programming skills in python is a must. Working experience with pandas, numpy, matplotlib, sklearn
• At least 3 years of industry experience in one or more full-time Data Science/ML roles
• Data Visualization techniques. Knowledge of visualization tools