AI Data Engineer

Key Responsibilities:

- Design, build and operate scalable data ingestion pipelines across a range of sources, including web data collection, API integrations and batch file deliveries, on our cloud data platform

- Build and productionise AI pipelines that use large language models (LLMs) and machine learning to extract, classify, enrich and standardise structured and unstructured data

- Develop agentic and retrieval-augmented generation (RAG) workflows, including prompt design, tool and Model Context Protocol (MCP) integrations and orchestration, where they deliver measurable gains in coverage, speed or quality

- Establish evaluation and monitoring for AI components, including accuracy benchmarks, human-in-the-loop review, drift detection, and cost and latency tracking

- Embed data quality, data governance and Responsible AI practices into every pipeline, in line with Dun & Bradstreet's AI Ethics and data management policies

- Research and apply modern data mastering and entity resolution techniques, including AI-assisted matching, to increase derived insight from disparate data sources

- Own the operational health of ingestion and AI pipelines in production, including monitoring, incident resolution and continuous optimisation of performance and cost

- Contribute to and help lead the AI & Ingestion community of practice by sharing reusable components, patterns and standards, running knowledge-sharing sessions and supporting the upskilling of engineers across teams

- Support other engineers through design reviews, code reviews and mentoring

- Develop and maintain documentation, including data flow diagrams, data lineage, pipeline runbooks, and prompt and model documentation

- Partner with product, data science and business stakeholders to translate data requirements into effective ingestion and AI solutions

Key Requirements:

- 5+ years of experience in data engineering or a similar position, including hands-on experience building data ingestion pipelines

- Strong Python and SQL are minimum requirements

- Practical experience building pipelines or applications with LLMs (for example via Anthropic, OpenAI or Google Vertex AI / Gemini APIs), including prompt engineering, structured outputs and evaluation

- Experience with cloud data processing, GCP preferred (e.g. BigQuery, Dataflow, Cloud Composer/Airflow, Vertex AI); Databricks experience is an advantage

- Experience in various methods of data ingestion and collection, such as web data collection, APIs, SFTP and other file delivery channels

- Familiarity with RAG, embeddings and vector databases, agent frameworks or MCP is desirable

- Understanding of engineering best practice for AI pipelines (MLOps/LLMOps), including version control, CI/CD, automated testing and monitoring

- Awareness of Responsible AI principles and of data privacy and governance considerations (e.g. GDPR) when applying AI to data

- A genuine passion for knowledge sharing, with experience of, or strong interest in, contributing to communities of practice, internal tech talks or mentoring

- Show an ownership mindset in everything you do; be a problem solver, be curious and be inspired to take action, be proactive, seek ways to collaborate and connect with people and teams in support of driving success.

- Continuous growth mindset: keep learning in a fast-moving AI landscape through social experiences and relationships with stakeholders, experts, colleagues and mentors, and widen your competencies through structured courses and programmes.

- Where applicable, fluency in English and languages relevant to the working market

- Robust data knowledge preferably around business data sets with proven analytical and quantitative skills

- Bachelor’s Degree in Data Science field, Information Systems, Computer Science, Mathematics or similar