5 of 5 Feature Engineering Jobs in the City of London

capable of reasoning and planning. Expertly handle LLM fine-tuning (PEFT, LoRA, QLoRA) and RAG pipelines. Data Orchestration: Build ETL/ELT pipelines and feature engineering workflows to integrate structured and unstructured data into centralized platforms. End-to-End MLOps: Own the lifecycle—from CI/CD automation … every system is trustworthy, fair, and explainable, implementing quantifiable metrics for bias detection and regulatory compliance. Technical Toolkit Models: LLMs, Generative AI, Agentic workflows. Engineering: PEFT, Vector Databases (Pinecone/Milvus/Weaviate), Prompt Engineering. Ops: Docker, Kubernetes, CI/CD, Experiment Tracking (MLflow/W&B). Data ...

governance frameworks to ensure responsible AI/ML model deployment. Use Python to automate processes and optimise workflows. Handle large datasets, including data preprocessing, feature engineering, and model evaluation. Lead the agile development of AI solutions from concept through deployment and ongoing improvement. Create and maintain comprehensive technical … documentation. Communicate complex AI concepts clearly to stakeholders at all levels. What you'll bring: Proven experience (3 years) in AI solution engineering, with a focus on large language models, prompt engineering, and RAG implementations. Strong skills in data analytics, MLOps practices, and API development. Knowledge of governance ...

user experience. What you'll be doing: → Design, train and optimise ML models for personalisation and recommendations → Build and maintain scalable data pipelines for feature engineering and model training → Deploy and monitor models in production — performance, availability, relevance → Lead A/B testing and offline experiments to drive ...

model Data Preprocessing: Design, build, and maintain robust ETL/ELT pipelines to ingest, transform, and load data from various sources. AI model Feature Engineering: Integrate structured and unstructured data from internal and external systems into centralized data platforms. Performance Tuning of AI/models: Optimize data workflows ...

experimentation strategy for the product area — designing A/B tests and multi-armed bandit experiments that generate genuine learning about AI feature performance, not just validate existing assumptions Drive the feedback loop strategy — defining how member behaviour is captured, modelled, and fed back into AI ranking systems … context, using results to change direction rather than confirm instinct Technically credible AI and ML fluency — able to read a model card, challenge a feature engineering decision, write testable acceptance criteria for ML features, and explain confidence intervals to non-technical stakeholders, without overstepping into engineering decisions ...