Senior Agentic Automation Engineer

EPAM is a leading global provider of digital platform engineering and development services. We are committed to having a positive impact on our customers, our employees, and our communities. We embrace a dynamic and inclusive culture. Here you will collaborate with multi-national teams, contribute to a myriad of innovative projects that deliver the most creative and cutting-edge solutions, and have an opportunity to continuously learn and grow. No matter where you are located, you will join a dedicated, creative, and diverse community that will help you discover your fullest potential.

• 3+ years of experience in software or automation engineering, with proven expertise in end-to-end agentic system design covering reasoning, decision-making, and execution
• Expertise in multi-agent architecture, including design patterns, cross-agent communication, and collaboration or critique between agents
• Proficiency in advanced context and memory techniques such as compression, multi-layer memory, retrieval/forgetting strategies, vector stores, and RAG pipelines
• Background in custom RAG and retrieval optimization focused on relevance, latency, throughput, and semantic search (dense embeddings, hybrid search, re-ranking)
• Knowledge of MCP at depth, including transport layers (HTTP, WebSocket, gRPC) and authentication (OAuth, token-based auth)
• Skills in scripting with JavaScript and/or Python to extend platforms and preprocess data
• Competency in designing secure, auditable pipelines with encryption, access control, PII handling, and traceability
• Familiarity with cloud agent frameworks (AWS, GCP, Azure) at an intermediate level
• Understanding of observability practices, including LLM-specific tooling, OpenTelemetry, and infrastructure monitoring with cost and performance tracking
• Capability to build reusable templates and produce cross-team documentation that enables broader adoption
• Fluency in English, both written and spoken, at a minimum B2 level

• Lead design and deployment of end-to-end agentic systems that combine reasoning, decision-making, and autonomous execution
• Design multi-agent architectures by applying established agent design patterns and robust cross-agent communication so agents can collaborate or challenge each other to improve outcomes
• Build advanced context and memory capabilities, including dynamic context compression, persistent multi-layer memory (short-term, long-term, episodic, shared), retrieval and forgetting strategies, and supporting infrastructure (vector stores, RAG pipelines, structured logs)
• Develop custom RAG and vector-based retrievers focused on retrieval relevance, latency/throughput optimization, and advanced semantic search (e.g., dense embeddings, hybrid search, re-ranking)
• Orchestrate tools via MCP with a solid grasp of transport layers (HTTP, WebSocket, gRPC) and authentication (OAuth, token-based auth)
• Script in at least one language supported by the platform (e.g., JavaScript, Python) to extend functionality or preprocess data
• Architect secure, auditable pipelines with encryption, access control, PII sanitization, and traceability so that actions can be reviewed and audited
• Work with cloud agent frameworks (e.g., AWS, GCP, Azure) at an intermediate level to design and operate solutions
• Integrate end-to-end observability using LLM-specific observability tooling, OpenTelemetry-based tracing, and infrastructure monitoring to track reliability, performance, and cost of agents and their tools
• Create reusable workflow templates and documentation so other teams can adopt and run agentic automation
• Own the full lifecycle of agentic solutions, from design and implementation to observability and iteration
• Enable others through clear documentation, templates, and guidance so the rest of the team can build on your work