Senior AI Engineer / Data Scientist

We're looking for a Senior AI Engineer / Data Scientist to join our team in London, United Kingdom, in a hybrid working mode. This role is at the core of EPAM’s Data & AI Practice and focuses on building state-of-the-art Generative AI, Agentic AI, and advanced data science solutions for real-world business problems. You’ll design and develop multi-agent systems, implement RAG pipelines, and deliver production-ready AI applications that transform client capabilities across multiple industries. If you're passionate about pushing the frontiers of LLMs, orchestration frameworks, and scaling AI systems into production, this opportunity offers a high-growth environment and cutting-edge challenges. Responsibilities Design, build, and deploy Generative AI and Agentic AI solutions from prototyping through production Develop and optimize multi-agent systems using frameworks such as LangGraph, CrewAI, AutoGen, and Semantic Kernel Implement orchestration patterns including planner/executor, supervisor/worker, and tool-calling workflows Design and build RAG pipelines, including embeddings, chunking, hybrid search, and retrieval evaluation for enterprise data grounding Develop orchestration engines supporting multi-step planning, delegation, and fallback paths for agent workflows Implement integration and communication patterns via MCP, A2A, OpenAPI, REST, and gRPC Build production-grade Python APIs and microservices integrating with enterprise systems and AI services Apply observability and monitoring solutions (Langfuse, Arize, Grafana) to ensure system reliability Contribute to solution architecture, best engineering practices, and documentation Requirements Bachelor’s/Master’s in Computer Science, Data Science, or related field with 4+ years’ experience, or Ph.D. with relevant experience Strong engineering experience with Python, APIs, microservices, debugging, and code review Proven experience building and deploying Generative AI or Agentic AI applications in production Deep understanding of LLM concepts, RAG patterns, prompt design, and evaluation methodologies Experience with multi-agent orchestration frameworks (LangGraph, CrewAI, AutoGen, Semantic Kernel) Familiarity with orchestration strategies like planner/executor and tool calling Knowledge of MCP, A2A protocols, and OpenAPI-based integration methods Strong experience with cloud environments, ideally Azure (Azure OpenAI, AI Foundry, AI Search) Competence in containerized deployments, CI/CD, and MLOps tooling (MLFlow, Airflow) Nice to have Experience with Microsoft Agent Framework, Azure AI Agent Service Knowledge of vector databases (Pinecone, Weaviate, Qdrant, Milvus) Familiarity with guardrail and AI safety techniques (output filtering, prompt injection defense) Experience in distributed systems, event-driven architectures, and workflow engines Prior involvement in training, fine-tuning, or experimenting with foundation models