Note: The job is a remote job and is open to candidates in USA. Cotality is a property intelligence company focused on making the property industry faster, smarter, and more people-centric. The Senior Geospatial Data Scientist will advance the company’s enterprise geocoding product by developing address parsing, geospatial precision, validation, benchmarking, quality scoring, and data licensing capabilities. The role also involves experimentation with modern geospatial tools and collaboration with engineering teams to bring features into production.
Responsibilities
- Complex Parsing: Build robust systems to handle the nuances of US addressing, including rural routes, high-rise secondary units, fractional addresses, non-postal locations, multiple valid address representations, and vanity address components
- Precision Engineering: Innovate new mechanisms to transition geolocation precision from areal (e.g., Postal, Census centroids) to specific (e.g., Parcel, Rooftop centroids) to differentiated (e.g., Front Door, Loading Dock, Drop-Off Points, etc.)
- Spatial Verification: Develop validation loops using multiple contributor data sources to verify that geocoded points link to other terrestrial datasets which link back to the original geocoded points
- Spot Verification: Evaluate and benchmark third-party geocoding services (e.g., Google, Mapbox, Overture Maps, etc.) vs. in-house builds to validate consistency and/or differentiation
- Quality Frameworks: Evolve automated “confidence” scoring systems to measure and report on the "health" of our addressing and geocoding data over time
- Ideate, iterate and implement enhancements to the geocoding stack to unlock higher accuracy, explainability and currency
- Perform hands-on experimentations using modern open-source, cloud native geospatial tools, and collaborate with engineering team to take features from idea to production
- Data Licensing Evaluations: Work with in-house data acquisition team to evaluate data vendors to ensure we are using best in class data while remaining competitive
- Evangelism: Drive awareness and education of address hygiene internally within the company as well as to externally to customers
Skills
- Mastery of USPS CASS/NCOA processes, TIGER data, and the nuances of the US postal system
- Experience of working with commercial as well as publicly available road networks datasets
- US land parcel data and relationship to addresses
- Census data
- 5+ years of relevant experience
- Be able to prototype and demonstrate ideas through hands-on coding in a Jupyter notebook oriented environment
- Python and SQL coding skills are a must
- Emphasis on geospatial and address and routing focused on Python libraries
- Geopandas, Shapely, Proj, Kepler/Pydeck
- Libpostal
- Valhalla
- Spatial data formats
- Legacy formats such as shapefiles
- Modern formats such as geojson, geoparquet, fgb
- PostGIS spatial SQL
- GDAL command line tools
- QGIS
- Working knowledge of REST APIs, JSON structures, PostMan
- Statistics fundamentals
- Bachelor's Degree or Equivalent Work History
- Some experience with a BigData stack such as Apache Spark
- Cloud based spatial analytics platforms such as Carto
- Geospatial pattern matching/recognition algorithms
Benefits
- Flexible working model
- Generous PTO and 11 paid holidays, plus well-being and volunteer time off.
- Up to 16 weeks of fully paid parental leave and a baby stipend.
- Multiple medical plan options with mental health and wellness support offerings.
- 401(k) with company match and vesting after one year.
- $400 annual well-being stipend and tuition assistance up to $5,250.
- Recognition Rewards, Referral bonuses, exclusive discounts and more!
Company Overview