Senior AI Engineer

We are looking for a Senior AI Engineer to design, build and productionise AI capabilities on an enterprise data and analytics platform. The role covers both generative AI (grounded conversational assistants using retrieval-augmented generation) and classical machine learning (forecasting, risk scoring, anomaly detection and root cause analysis), delivered as secure, scalable services on AWS.

You will own the AI/ML service end to end: retrieval and grounding design, model selection, evaluation harness, API integration with backend and front-end teams, and production readiness including guardrails, observability and access-scoped data handling. The role suits an engineer who has shipped AI features to real users and cares as much about accuracy, citation and verification as about model capability.

• Design and build a grounded conversational assistant using RAG: ingestion, chunking, embeddings, hybrid (vector + keyword) retrieval, reranking and citation of source records


• Implement LLM orchestration: query routing, tool/function calling against governed data APIs, response assembly and answer verification against retrieved evidence


• Build and maintain an LLM evaluation harness covering relevance, groundedness, citation accuracy, latency and regression testing


• Develop forecasting, risk scoring and anomaly detection models on high-volume time-series and tabular data


• Support AI-assisted root cause analysis: feature attribution, probability ranking of contributing causes and evidence summarisation


• Expose AI/ML capabilities as secure REST APIs (Python FastAPI) deployed as containers on Kubernetes (EKS)


• Integrate LLMs through private endpoints (e.g. AWS Bedrock / VPC endpoints) so data does not leave the controlled network


• Enforce role-based access and data scope in retrieval, prompts and conversation state


• Implement guardrails, graceful degradation, prompt/version management and audit logging


• Set up monitoring and observability for model and assistant performance; use production logs to drive continuous evaluation


• Work with data engineers on feature and predictive datasets, and with front-end and backend engineers on API contracts


• Participate in Agile/Scrum ceremonies, estimation and sprint delivery