Lead Agentic AI Engineer

EPAM Systems; Argentina; On-site; ARS 68,215,000 - 113,691,000; Full time

Benefits offered by this job

Healthcare benefits

Employee financial programs

Paid time off and sick leave

Upskilling, reskilling and certs

Job summary

EPAM Systems seeks a Lead Agentic AI Engineer to architect, deploy, and operate autonomous agent systems at scale for production AI deployments. You will guide decisions around runtime, memory, identity, tools, and observability using AWS Bedrock AgentCore and various agentic frameworks.

Collaborate with engineers to improve reliability, safety, and cost efficiency. Responsibilities include designing end-to-end architecture for agents, leading framework adoption, and building robust MCP servers

Qualifications

• 5+ years of production software experience with 2+ years shipping LLm-based agents to production.
• Proficient in Python and familiar with agent frameworks (Strands, LangGraph, Claude Agent SDK, OpenAI Agents SDK, CrewAI).
• Knowledge of AWS Bedrock AgentCore primitives (Runtime/Memory/Identity/Observability).
• Strong understanding of MCP, tool usage, prompts, and memory architectures.
• Design for production including observability, guardrails, retries, cost control, and security.

Responsibilities

• Design end-to-end architecture for autonomous agents across runtime, memory, identity, tools and observability.
• Lead adoption of AWS Bedrock AgentCore and agent SDKs in production deployments.
• Architect long-term memory and context-management for multi-step engagements.
• Build and curate MCP servers and tool integrations accessed by agents.
• Optimize reliability, cost, and latency of agent reasoning.
• Define standards for evals, tracing, and guardrails.

Skills

Python

Agentic frameworks

English Proficiency (B2+)

Production software

Observability

Memory architectures

Tools

Strands

LangGraph

Claude Agent SDK

OpenAI Agents SDK

CrewAI

AWS Bedrock AgentCore

Job description

We are seeking a Lead Agentic AI Engineer to architect and run autonomous agent systems at scale for production AI deployments. You will guide decisions across runtime, memory, identity, tools, and observability using AWS Bedrock AgentCore and agentic frameworks. Partner with engineers to raise reliability, safety, and cost efficiency.

Responsibilities

• Design end-to-end architecture for building, deploying, and operating autonomous agents across runtime, memory, identity, tools, and observability
• Lead adoption of AWS Bedrock AgentCore and the Strands Agents SDK across agent harnesses and production deployments
• Architect long-term memory and context-management mechanisms that remain robust across multi-step engagements
• Build and curate MCP servers and tool integrations agents call, including recon, scanners, exploit primitives, and custom internal services
• Optimize reliability, cost, and latency of agent reasoning
• Define standards for evals, tracing, and guardrails

Requirements

• Proven experience building production software for 5+ years, including 2+ years delivering LLM-based agents into production
• High proficiency in Python plus hands-on use of an agentic framework such as Strands, LangGraph, Claude Agent SDK, OpenAI Agents SDK, or CrewAI
• Working knowledge of AWS Bedrock AgentCore primitives (Runtime, Memory, Identity, Gateway, Observability) or demonstrated ability to ramp up quickly
• Solid understanding of MCP, tool use, prompt/context engineering, memory architectures, and agent evaluation
• Demonstrated ability to design for production, including observability, guardrails, retries, cost control, and security boundaries
• English proficiency at B2 (Upper-Intermediate) level or higher

Nice to have

• Background in application security, web pentesting, or red-teaming
• Experience with sandboxed code execution, browser agents, or autonomous tool orchestration

We offer

• International projects with top brands
• Work with global teams of highly skilled, diverse peers
• Healthcare benefits
• Employee financial programs
• Paid time off and sick leave
• Upskilling, reskilling and certification courses
• Unlimited access to the LinkedIn Learning library and 22,000+ courses
• Global career opportunities
• Volunteer and community involvement opportunities
• EPAM Employee Groups
• Award-winning culture recognized by Glassdoor, Newsweek and LinkedIn