The primary duty of a Lead Scientist is to research, evaluate, and implement advanced descriptive and predictive data models, with a preference for unsupervised techniques based on Statistical Analysis, and Graph-based methods. Familiarity/Expertise with Deep Learning, Unstructured Data Analysis, NLP as well as extensive software engineering skills is required.
What you will contribute:
- Research, evaluate, and build descriptive and predictive decision strategies to be implemented as part of our FICO Cognitive Analytics pipeline. In addition to familiarity with classical data analysis tasks, such as data cleansing, wrangling, pattern identification and variable creation, the position will focus on the use of new techniques for data segmentation and analysis to bring innovation to our cognitive analytics group. Examples of the later are research and implementation of anomaly detection and outlier detection methods based on graph analytics and connected data analysis.
- Lead technical implementation of models and innovative solutions for new analytics in fraud detection, health care, and decision recommendation.
- Collaborate tightly with an existing team of FICO Scientists doing applied R&D on cognitive analytics, while actively contributing with the engineering team tasked with the implementation of these new analytic approaches. Keep up-to-date with latest technologies trends. Communicate results and ideas to the rest of the team.
- Assist with client meetings and client engagements to promote the new technology, implement proof of concepts, and solve real life data problems. Provide support for customer meetings, model construction, and pre-sales engagements.
- Write documentation related to research and product features to present to both technical and non-technical audiences.
- Serve as a source of technical expertise and leadership across development teams.
- Travel required 10% of the time.
What we’re seeking:
Data Science Skills
- Master’s Degree or Ph.D. (preferable) in Computer Science, Statistics, Electrical Engineering, Applied Math, Physics, or related field.
- Experience with data pre-processing, and data analytics. Extensive background in Data Mining and Statistical Analysis. Able to understand various structures and common methods of data transformation.
- Strong pattern recognition and predictive modeling skills.
- Experience with different supervised and unsupervised machine learning methods for clustering, classification, and forecasting. Linear, Non-Linear and Logistic Regressions, Neural Networks, SVMs, Random Forests, and Outlier Detection methods.
- Experience with deep learning networks such as Long-Short Term Memory (LSTM) and Convolution neural Networks (CNN), as well as Attention Based Modeling and Embeddings.
- Familiarity with Graph-based algorithms highly desirable.
- Ability to code in Java, Phyton, C, C++. Scripting experience using Perl, Bash, and Linux to facilitate data analysis and model development.
Software Engineering Skills
Tools and Frameworks:
- Python: ML frameworks (Tensorflow, Scikit-Learn, etc.), Flask, Requests, Tenacity, SpaCy, Datasketch, PyTest, opencv-python, etc.
- Java: Spring framework, Guice, Jersey, SLF4J, Logback, JUnit, etc.
- C++: OpenCV, Tesseract, Boost, Pistache
- MongoDB, MySQL, ArangoDB, Oracle
Development, Build and CI/CD tools:
- Git, Maven, Gradle, Pantsbuild, Jenkins, Sonar, Nexus, Helm
- OAuth2, OpenID Connect, SAML2