The posting, in Farfetch's own words
archived Oct 19, 2026WHAT YOU'LL DO:
Partner with stakeholders across Catalogue and Content, Product and Data to understand business problems, challenge existing processes, identify automation opportunities and translate them into practical solutions, from straightforward process automation to more complex AI-driven solutions; Own the full development lifecycle from initial problem framing and prototyping to production deployment, monitoring and ongoing optimization, including system design, testing, evaluation and continuous improvement; Build robust GenAI applications and LLM-based workflows leveraging RAG, structured outputs, function/tool calling, context management and multi-step reasoning workflows where appropriate; Construct and orchestrate reliable backend services and pipelines using Python, Apache Airflow, APIs and cloud-native solutions, optimizing for system reliability, latency and inference/token costs; Define and track operational metrics (throughput speed, error reduction, operational hours saved, metadata quality) to measure the tangible ROI of every deployed solution; Establish reusable patterns, practices and capabilities for building reliable, maintainable and scalable AI and automation solutions, implementing AI safety best practices (prompt injection defense, guardrails, output validation); Explore and prototype emerging AI capabilities, evaluating how they can solve real business problems and evolve into production-ready solutions; Help shape and expand our AI practice into a fully formed capability and service offering for the Catalogue and Content domain, combining systems design, consulting, delivery, and innovation.
Must have:
5+ years of relevant development experience (Software Engineering, MLOps, Applied AI, Automation or Data); A pragmatic approach to technology: knowing when to use AI, conventional automation or neither; Deep expertise in Python, SQL and designing clean, maintainable backend services/microservices; Hands-on experience with workflow orchestration platforms (e.g. Apache Airflow) and webhooks/API integrations; Strong practical knowledge of prompt engineering, vector databases/embeddings, RAG architectures and output validation frameworks (e.g., Pydantic, Instructor, Guardrails); Familiarity with modern software delivery workflows: version control (Git), containerized applications (e.g., ) and basic cloud deployments; Demonstrated experience with AI safety considerations: prompt injection defense, output validation and reliability engineering in production; Ability to work across technical and business contexts and communicate effectively with non-technical stakeholders.