Tegumen is an international risk consultancy delivering management consulting services and developing risk management technology.
We are looking for a junior data science/engineering students to help with a multitude of data collection and analytics tasks associated with an ambitious and high-impact technology startup based in Miami, FL. Opportunities include both paid and unpaid positions aimed at offering hands-on experience with a recognized risk mitigation consulting team and a high-potential startup venture. Time commitment may range between ten and 20 hours/week, over two to four months. Through these internship and part-time/short-term consulting assignment opportunities, we hope to identify highly motivated and competent candidates to join the core team, potentially in more permanent roles.
Our work and products focus on the domains of individual risk, including criminal victimization, safety & transportation, health & disease, etc. We are constantly exploring, ingesting and analyzing data sets in these domains to produce new insights, find patterns and trends, visualize data for our team or customers, and develop our proprietary algorithms.
The data you’ll be working with will include, but not be limited to: Centers for Disease Control (CDC), the FBI’s Uniform Crime Report (UCR), National Crime Victimization Survey (NCVS), Global Health Data Exchange (GHDx), and other.
The work you'll be expected to perform includes:
- Collect raw data collection (e.g., scraping, APIs )
- Perform basic data engineering/ETL (e.g., data cleaning, transformation, and database design)
- Develop interactive visualizations (preferably using open source libraries)
- Apply machine learning and parametric statistical modeling (primarily risk models)
- Work with software developers to improve the user interface and data representation so as to better execute our core algorithms
- Assist with formulating extensions to the core algorithm and improving imputation techniques
- Prototype and present solutions to company leadership
Our ideal candidate will have the following skills/competencies:
- Fluent in English
- Advanced statistics & probability knowledge, particularly in the areas of epidemiological and/or actuarial modeling
- Experience with Bayesian modeling tools (e.g., Stan) and their idiosyncrasies is a plus
- Familiarity with basic SQL, main database types, and basic data modeling
- Solid R and/or Python programming ability -- ability to develop and debug analytical scripts with minimal guidance and supervision
- Knowledge of geospatial analysis and geostatistical models (e.g., gaussian processes a.k.a. kriging) is a big plus, preferably performed using scalable, open-source tools such as PostGIS and QGIS