Senior AI Engineer: Agent Orchestration & HITL Lead

EPAM Systems; Turkey; On-site; TRY 400,000 - 700,000; Full time

Benefits offered by this job

English courses taught by certified老师s

Private health insurance

Continuous upskilling and development

Job summary

EPAM Systems, Inc. is seeking a Senior AI Engineer to own the Agent Orchestration Framework within our Enterprise Agent Development Platform. You will design multi-agent orchestration patterns, implement checkpointing/state management, and architect HITL workflows for enterprise approvals.

You will work with LangGraph, LangChain, and AWS Bedrock to deploy scalable agent runtimes, collaborating with multiple teams to standardize practices and improve observability and reliability in production.

Qualifications

• 5+ years of Python engineering experience.
• Hands-on experience with LangGraph and LangChain for multi-step agent workflows.
• Experience with AWS Bedrock AgentCore Runtime.
• Experience with workflow orchestration tools such as Temporal or Airflow.

Responsibilities

• Own the design and evolution of the Agent Orchestration Framework across the platform.
• Design and implement multi-agent orchestration patterns (supervisor/worker, fan-out/fan-in).
• Build and maintain checkpointing and state management for LangGraph workflows.
• Architect HITL workflows to support enterprise approvals.
• Enable A2A agent delegation via the Agent Gateway.
• Develop and maintain agent runtimes on AWS Bedrock AgentCore Runtime.
• Collaborate with teams to standardize agent development practices and ensure observability, reliability, and scalability in production.

Skills

Python

LangGraph

LangChain

Agent orchestration

HITL workflows

Distributed state management

AWS Bedrock

Workflow orchestration

Tools

Temporal

Airflow

Bedrock

Job description

EPAM Systems, Inc. is seeking a Senior AI Engineer to own the Agent Orchestration Framework within our Enterprise Agent Development Platform. You will design multi-agent orchestration patterns, implement checkpointing/state management, and architect HITL workflows for enterprise approvals.

You will work with LangGraph, LangChain, and AWS Bedrock to deploy scalable agent runtimes, collaborating with multiple teams to standardize practices and improve observability and reliability in production.