Remote Lead, AI-Native Systems & Agentic Architecture

EPAM Systems; Georgia; Hybrid; USD 64,000 - 90,000; Full time

Benefits offered by this job

Engineering Heritage

Advanced Tech Stack

World-Class Clients

Professional Growth

GenAI Community

Hybrid Latvia

Other Benefits

Job summary

EPAM Systems Latvia is seeking an experienced engineer to build AI-native systems end-to-end, including agents, workflows, and RAG harnesses. You will work with SMEs and end-users, shaping use cases and validating technical feasibility in real projects.

This role requires 7+ years in engineering, strong Python skills, hands-on with LangChain/LangGraph and major LLM providers, plus production deployment on cloud and observability tools. English at C1 and a client-facing mindset are essential.

Qualifications

• 7+ years of engineering experience in production AI/LLM applications.
• Strong agent-design judgment and task-harness fit.
• Ability to lead discovery with SMEs and end-users and explain trade-offs.
• Hands-on with LangChain, LangGraph and Semantic Kernel; with major LLM providers.
• Expert-level Python and solid software engineering fundamentals.
• Strong RAG and retrieval skills; experience with evaluation and observability.
• Production cloud deployment experience with containerization and CI/CD.
• Clear, confident English communication (C1).

Responsibilities

• Design, build and ship AI-native systems end-to-end: agents, workflows, RAG and harness.
• Build evaluation pipelines to prove real usefulness of the system.
• Design for failure in agent loops, with retries, fallbacks and human-in-the-loop.
• Capture domain expertise and create repeatable workflows for future engagements.
• Engage early with clients, shape use cases and assess technical feasibility.
• Collaborate with SMEs and end-users through interviews, UAT and observation.

Skills

AI/LLM engineering

Agent design

Client engagement

LangChain

Semantic Kernel

Python

RAG retrieval

LLM evaluation

Cloud deployment

Tools

LangSmith

Arize Phoenix

Langfuse

AWS/Azure/GCP

Job description

EPAM Systems Latvia is seeking an experienced engineer to build AI-native systems end-to-end, including agents, workflows, and RAG harnesses. You will work with SMEs and end-users, shaping use cases and validating technical feasibility in real projects.

This role requires 7+ years in engineering, strong Python skills, hands-on with LangChain/LangGraph and major LLM providers, plus production deployment on cloud and observability tools. English at C1 and a client-facing mindset are essential.