Lead Python AI Engineer

EPAM Systems is hiring a Lead Python AI Engineer to build robust AI-powered API services and agentic workflows. The role focuses on secure, scalable LLM integrations, backend services, RAG pipelines, and reusable orchestration frameworks, while contributing to engineering standards and collaborating across teams.

Responsibilities:

• End-to-End Ownership: Own API contracts and service reliability end-to-end. AI edge cases must fail gracefully without compromising system integrity
• Rigorous Systems Engineering: Integrate LLMs with a strong engineering foundation—baking in evaluation, comprehensive observability, robust fallback handling, and cost awareness
• Reusable Frameworks: Build reliable, reusable orchestration frameworks and logic designed for downstream developer consumption
• Design and build scalable backend API services and advanced LLM orchestration layers
• Implement and maintain complex RAG (Retrieval-Augmented Generation) pipelines, covering document ingestion, chunking, embedding, and retrieval optimization
• Develop and connect agent tools using LangChain, LangGraph, and potentially MCP (Model Context Protocol)
• Ensure top-tier security, privacy, enterprise-grade observability, and comprehensive test coverage across all backend workflows
• Contribute actively to architecture decisions and engineering standards within your pod
• Collaborate cross-functionally with frontend engineers, data engineers, and infrastructure teams

Requirements:

• 7+ years of backend engineering with a heavy focus on microservices and distributed systems, using Python for high-performance backend services and cloud-native APIs
• Hands-on experience with AWS, Docker, ECS/EKS, and RESTful APIs
• Proven experience with secure coding practices and strong authentication/authorization (auth/authz) fundamentals
• Practical, hands-on experience with frameworks such as LangChain, LlamaIndex, or LangGraph
• 1–2+ years of hands-on experience integrating AI SDKs (OpenAI, Anthropic/Claude, AWS Bedrock, or equivalent) in production environments
• Familiarity with AI evaluation tooling or real-time APM platforms (e.g., LangSmith, Langfuse, Arize)
• Practical experience working with vector databases (e.g., Amazon Kendra, OpenSearch), advanced embedding strategies, and retrieval systems
• 1–2+ years of experience with React and TypeScript
• Experience deploying applications to EKS environments at scale
• Knowledge of agent interoperability patterns (MCP) and identity/security domain models (IAM, CIAM)
• Professional proficiency in Java, Node.js, or Go