Senior Agentic AI Engineer

EPAM Systems; Argentina; Remote; ARS 136,660,000 - 227,766,000; Full time

Benefits offered by this job

Healthcare benefits

Paid time off

Upskilling and certs

LinkedIn Learning access

Global career opportunities

Job summary

EPAM Systems, Inc. is seeking a Senior Agentic AI Engineer to architect, build, and run autonomous agent systems at scale. You will make key architectural decisions spanning runtime, memory, identity, tools, and observability for production AI systems.

You will lead adoption of AWS Bedrock AgentCore and Strands/LangGraph frameworks, develop memory and context-management, and improve reliability and cost efficiency of agent reasoning. Strong Python is required.

Qualifications

• 3+ years building production software with at least 1 year deploying LLM-based agents.
• Strong Python and hands-on experience with agentic frameworks.
• Knowledge of AWS Bedrock AgentCore components and related tooling.
• Solid grasp of memory, prompt engineering, and agent evaluation methods.
• English proficiency at least B2 level.

Responsibilities

• Architect autonomous agent systems spanning runtime, memory, identity, tools, and observability.
• Develop MCP servers and tool integrations that agents rely on.
• Improve reliability, cost-efficiency, and speed of agent reasoning.
• Establish standards for evaluations, tracing, and guardrails.

Skills

Python

Agent frameworks

LLM deployment

Observability

Memory architecture

Tools

Strands

LangGraph

Claude SDK

OpenAI Agents SDK

CrewAI

AWS Bedrock AgentCore

Job description

We're looking for a Senior Agentic AI Engineer to architect, build, and run autonomous agent systems at scale, making key architectural decisions spanning runtime, memory, identity, tools, and observability for production AI systems.ResponsibilitiesArchitect systems for building, deploying, and running autonomous agents, spanning runtime, memory, identity, tools, and observabilitySpearhead adoption of AWS Bedrock AgentCore and the Strands Agents SDK for agent harnesses and production rolloutsCreate long-term memory and context-management systems capable of sustaining multi-step engagementsDevelop and maintain MCP servers and tool integrations that agents rely on, such as recon tools, scanners, exploit primitives, and internal custom servicesImprove the reliability, cost-efficiency, and speed of agent reasoningEstablish standards for evaluations, tracing, and guardrailsRequirements3+ years building production software, with at least 1 year deploying LLM-based agents into productionStrong Python skills and hands-on experience with agentic frameworks like Strands, LangGraph, Claude Agent SDK, OpenAI Agents SDK, or CrewAIKnowledge of AWS Bedrock AgentCore components—Runtime, Memory, Identity, Gateway, and Observability—or the ability to learn them quicklySolid grasp of MCP, tool use, prompt/context engineering, memory architectures, and agent evaluation methodsSkill in designing production-ready systems, including observability, guardrails, retry logic, cost management, and security boundariesEnglish proficiency at B2 level or higherNice to haveExperience in application security, web pentesting, or red-teamingBackground with sandboxed code execution, browser agents, or autonomous tool orchestrationWe offerInternational projects with top brandsWork with global teams of highly skilled, diverse peersHealthcare benefitsEmployee financial programsPaid time off and sick leaveUpskilling, reskilling and certification coursesUnlimited access to the LinkedIn Learning library and 22,000+ coursesGlobal career opportunitiesVolunteer and community involvement opportunitiesEPAM Employee GroupsAward-winning culture recognized by Glassdoor, Newsweek and LinkedIn