Streets Data Scientist

To improve the quality, consistency and analytical value of the National Street Gazetteer (NSG) - the authoritative geospatial dataset of highways in England and Wales.

The post holder supports local highway authorities in the submission and incorporation of street data from Local Street Gazetteers (LSGs) and third-party suppliers, and applies data science techniques at national scale: profiling authority submissions, detecting errors, investigating root causes, and building checks to make it simple for custodians to improve the quality of their data.

This is a hands-on technical role. Roughly half the work is direct custodian support and error resolution; the other half is analysis, tooling and automation that bridges the gap between the work of the Data Insights team and the ELT Pipeline Developers., Custodian support and data resolution

• Act as a first technical point of contact for LSG custodians in response to files submitted to GeoPlace.
• Supporting LSG custodians with compliance errors, health check error resolution and exceptions.
• Monitoring the Annual Improvement Schedule levels and proactively contacting custodians.
• Monitoring and reporting of system bugs.
• Represent GeoPlace at Regional Custodian Group meetings in England and Wales, * Profile street data to quantify quality: completeness, currency, internal consistency, and conformance to the Data Transfer Format and NSG Data Entry Conventions.
• Design and implement automated data quality checks that run across all contributing authorities, including statistical and rule-based anomaly detection to surface unusual submissions - sudden volume changes, geometry churn, orphaned or duplicated records, implausible attribute distributions.
• Assist with other data loads (for example the OS MasterMap Highways Network), including conflation and matching of external network data against the NSG.
• Support system testing and user assurance testing, including the design of test datasets and automated regression checks against known data conditions.
• Assisting with system testing as required.

Other Responsibilities

• Provide technical advice and content to the communications team, contributing analysis and evidence to case studies and blogs relating to the NSG.
• Ad hoc spatial analysis in support of GeoPlace, its stakeholders and wider government initiatives.
• Contribute to the documentation of data models, business rules and analytical methods so that work is transparent and repeatable by others., The post holder will undertake other duties as appropriate to the post to support and contribute to the delivery of job role and the GeoPlace Vision, Mission and Objectives as described on the GeoPlace website.

• SQL -confident writing and optimising queries against large relational datasets.
• Data quality analysis and root-cause investigation -able to move from a symptom in a dataset to a defensible explanation of why it occurred.
• Analytical and statistical thinking - comfortable summarising distributions, identifying outliers, and judging whether a change in the data is meaningful or noise.
• Communication -able to explain technical findings clearly to technical and non-technical audiences alike, in writing and in conversation, including to external authority staff.
• Attention to detail, combined with the judgement to know which anomalies matter.
• GIS tools such as QGIS or ArcGIS, and familiarity with geospatial libraries and formats (GeoPackage, Shapefile, GML, GeoJSON).
• Experience of at least one of the UK address and street data standards -BS 7666, the NSG Data Transfer Format, USRNs and UPRNs, OS MasterMap Highways Network, NGD Transport Street Manager

Desirable

• Python or R for data manipulation, spatial analysis and automation.
• Oracle Spatial, including spatial indexing and spatial SQL functions.
• Linear referencing and network data -experience with linear referenced networks, routing, or topology validation.
• Machine learning or predictive modelling applied to data quality, classification, matching or change detection.
• Version control and reproducible workflows -Git, scripted and documented analysis, testing.
• Good knowledge of Highways Legislation including NRSWA, HA and TMA, * A minimum of one year’s experience working with spatial data in a professional or academic setting.
• Demonstrable experience using SQL scripting against real datasets.
• Experience of GIS or other spatial data tools.
• Experience of working with, or within, local government, highways, utilities, transport or a comparable data-intensive public sector domain is an advantage, but not a requirement.

Please note: This role is based at our central London office, with the ability to work from home. Some regular business travel across England and Wales is also required. During your onboarding and probation period, we’d like to agree a workable office visit schedule with you, ideally averaging around one day a week, to support you in settling into the role

• 25 days holiday (in addition to normal and public bank holidays) per year
• 10% of your working week can be spent taking part in Learning and Development
• Group Personal Pension (minimum employer contribution is 4%)
• Employee subscription to Udemy training
• Group Income Protection Scheme (on completion of probation)
• Life Assurance Scheme - 3 x salary (on completion of probation)
• Interest free season ticket loan (on completion of probation)
• Cycle to Work scheme (on completion of probation)