AI-Native Systems Engineer

EPAM Systems, Inc.; Turkey; On-site; TRY 5,395,000 - 8,828,000; Full time

Job summary

EPAM Systems, Inc. is seeking an experienced AI/LLM engineer to design and build agentic systems end-to-end. You will craft tool calls, harnesses, and evaluation pipelines, collaborating with SMEs and end-users to validate feasibility and value.

You will write production-grade Python, enable scalable deployments on major clouds, and implement robust failure handling with human-in-the-loop when needed. Strong communication across technical and non-technical audiences is essential.

Qualifications

• 7+ years of engineering experience with a strong recent track record building production AI/LLM apps.
• Proven agent-design judgment — task-harness fit and policy-aware implementations.
• Experience deploying in production on major cloud platforms with containerization and CI/CD.
• Excellent ability to explain technical trade-offs to technical and non-technical stakeholders.

Responsibilities

• Design, build and ship AI-native systems end-to-end, including agents and harnesses.
• Develop evaluation pipelines and prove usefulness with real workflows.
• Write production-grade Python for integrations, APIs, and data access.
• Collaborate with SMEs and end-users to shape use cases and feasibility.
• Implement robust error handling, retries, and human-in-the-loop controls.
• Capture domain knowledge to reproduce success across projects.

Skills

Agent design

Python programming

LangChain

LangGraph

Semantic Kernel

RAG / retrieval

Observability / tracing

CI/CD

Cloud fundamentals

Communication with SMEs

Education

NLP / Data Science background

Tools

LangSmith

Arize Phoenix

Langfuse

AWS Bedrock AgentCore

Databricks Genie

Microsoft Foundry

Job description

EPAM Systems, Inc. is seeking an experienced AI/LLM engineer to design and build agentic systems end-to-end. You will craft tool calls, harnesses, and evaluation pipelines, collaborating with SMEs and end-users to validate feasibility and value.

You will write production-grade Python, enable scalable deployments on major clouds, and implement robust failure handling with human-in-the-loop when needed. Strong communication across technical and non-technical audiences is essential.