Appearance
Hélène Draux
Senior Data Scientist & Technical Data Consultant
Senior data scientist and technical consultant with 9+ years' experience delivering complex analytical projects across data infrastructure, AI-enabled analytics, and research intelligence. I build governed analytical systems: the method can be reconstructed, the assumptions challenged, and the output defended.
I have repeatedly moved into new technical domains quickly, from spatial data and research methods to large-scale research analytics, BigQuery-based consulting, and LLM-mediated analytical systems, building enough depth to deliver client-facing work, reusable frameworks, and senior stakeholder communication.
CV of Hélène Draux, Senior Data Scientist at Digital Science, London — available for senior roles and consulting in research intelligence, AI analytics, and bibliometric data science.
Experience
Senior Data Scientist (Thought Leadership), Digital Science, London
January 2024 – Present
Leading applied research and public-facing technical work on AI-assisted analytics, data governance, and research-intelligence infrastructure, with a focus on how LLMs can interact reliably with structured analytical data.
- Designed and built a hybrid AI classification pipeline (Python + SLMs + LLMs) to categorise research funder organisations across five analytical dimensions.
- Developed a governance framework for conversational analytics, formalising how semantic constraints, structured data context, and workflow enforcement can make LLM-mediated analysis more reliable, auditable, and commercially usable.
- Designed and prototyped an AI-assisted research-intelligence workflow translating natural-language analytical intent into structured analytical inputs over large-scale bibliometric data.
- Produced a multi-part technical series on AI-mediated analytics architecture, covering semantic governance, methodological authority, workflow enforcement, and the limits of prompt-only approaches.
- Contributed to internal best practice on how AI tools should interface with structured data platforms, balancing usability, reproducibility, and analytical validity.
Data Scientist (Consultancy), Digital Science, London
April 2017 – January 2024
Core member of the research data science team, owning projects end-to-end from problem definition through to client-facing delivery.
- Managed a portfolio of client projects involving requirements gathering, proposal writing, client discussions, analytical execution, and presentation of findings to technical and non-technical stakeholders.
- Built and maintained data models, classification systems, and analytical workflows over large-scale bibliometric datasets using Google BigQuery and related data infrastructure.
- Developed complex SQL pipelines on Google BigQuery, including multi-step classification workflows with statistical thresholds, normalisation, and multi-criteria labelling to surface emerging research trends across 22 fields.
- Created Looker dashboards and interactive visualisations to make complex analytical findings accessible to internal and client-facing audiences.
- Acted as the technical bridge between data infrastructure, analytical methodology, and client needs, translating constraints and trade-offs into practical recommendations.
Postdoctoral Researcher, Copenhagen University, Copenhagen
July 2013 – September 2016
- Led development of a Google Maps-based platform for geolocated data collection and a Leaflet/CartoDB/D3.js platform for public-facing results display.
- Designed and administered a spatially-enabled citizen science questionnaire from concept through deployment.
- Delivered methodology lectures in webGIS, outdoor recreation management, and human movement tracking.
Technical Consultant, Institut National de l'Information Géographique et Forestière, France
February 2011 – April 2013
- Advised public and private organisations on geographic data acquisition and use.
- Managed client relationships from initial meeting through order fulfilment and internal liaison.
Education
- PhD, Social Geography: London Metropolitan University, 2010–2013. Participatory GIS and land use planning; EU 7th Framework project (SECOA). Scholarship from Vice-Chancellor scheme.
- MSc, Land Management: Cranfield University, 2007–2008. Dual degree with ENSG; spatial data, environmental systems, GIS, and quantitative analysis applied to real-world land use contexts.
- Diplôme d'Ingénieur, Géomatique: ENSG (École Nationale des Sciences Géographiques), 2005–2008. French Grande École engineering degree. Geodesy, cartography, remote sensing, spatial databases, GIS, statistics, and scientific programming.
- Classe Préparatoire (PCSI): Lycée Descartes, Tours, 2002–2005. Intensive preparation in physics, chemistry, and mathematics for competitive Grande École entrance.
Expertise
- Data & Analytics: Python, SQL (advanced), Google BigQuery, Looker, D3.js, CartoDB, Dimensions
- AI & Pipelines: LLM/SLM orchestration, hybrid AI workflows, prompt engineering, conversational AI
- Methods: Data modelling, classification systems, bibliometrics, spatial analysis, GIS
Languages
French (native) · English (fluent)
Selected technical work
- Conversational Bibliometrics & MCP Governance: Authored a public technical series on the infrastructure of AI-mediated analytics, with direct relevance to governed semantic-layer design.
- Global Research Funder Classification: End-to-end hybrid AI pipeline: deterministic rules, SLM verification, LLM patching, and LLM-as-judge validation across 7,500 organisations (December 2025).