About this role
End Date Thursday 30 July 2026 We Support Flexible Working – Click here for more information on flexible working options
Flexible Working Options Hybrid Working Job Description Summary AI Engineer – Grade D Location: Hyderabad – Lloyds Technology Centre Function: Lending & Working Capital Platform Experience: 4–6 years (software/ML/AI); proven production delivery, hands on GenAI and Agentic AI experience
Location: Hyd Mode: Hybrid YOE: 4-6 Job Description Role Purpose Lead the design and delivery of enterprise-scale AI/ML solutions—including LLM/GenAI features—with strong focus on reliability, security, and compliance. Drive technical standards, mentor junior engineers, and collaborate with cross-functional teams to operationalise AI safely and efficiently. Key Responsibilities
• AI Solution Design & Delivery: Architect and implement advanced ML and GenAI systems; optimise for performance, cost, and scalability. • Model Operationalisation (MLOps): Build CI/CD pipelines, implement automated testing, and manage model lifecycle with MLflow or equivalent. • LLMOps & GenAI: Develop RAG workflows, embeddings, and vector indexes; enforce prompt safety, observability (latency, token usage, cost), and guardrails. • APIs & Integration: Expose models via secure microservices (FastAPI or similar); ensure RBAC/ABAC and audit logging. • Governance & Compliance: Embed AI ethics, regulatory standards, and security controls into all solutions. Essential Skills
• Strong Python and software engineering discipline; working knowledge of SQL. • Hands-on with Docker/Kubernetes and Git-based CI/CD (GitHub/Azure DevOps). • Experience with cloud AI stacks (GCP Vertex AI), artefact registries, and secrets management. • Deep understanding of LLM fundamentals (prompting, embeddings, RAG, guardrails). • Familiarity with MLflow/Kubeflow, Airflow/Composer, and feature stores (e.g., Feast). Desirable Skills
• Vector DBs (PGVector/Weaviate/Pinecone), LangChain/LlamaIndex. • Observability tools (Prometheus/Grafana/OpenTelemetry) and model evaluation frameworks (Evidently, Ragas/TruLens). • Secure engineering practices: tokenisation/masking, KMS/Key Vault, policy-as-code.