Senior AI-Native Engineer

EPAM Systems; United States; On-site; USD 140,000 - 190,000; Full time

Job summary

EPAM Systems Inc is seeking a Senior AI-Native Engineer who loves understanding complex systems, tracking stubborn bugs, and shipping AI-powered features customers actually use.

You will design and implement AI workflows, configure and extend agent harnesses, and collaborate with product and engineering teams to move faster while preserving quality and reliability.

You will work with real repositories, evaluate agent performances, and help teammates adopt scalable AI practices across projects.

Qualifications

• 5+ years shipping and maintaining production software.
• Experience writing code, reasoning about architecture, debugging unfamiliar systems, and explaining technical decisions.
• Daily experience with coding agents such as Claude Code, Codex, Cursor.
• Understanding of agent capabilities, limitations, and failure modes.
• Hands-on experience configuring and adapting agent harnesses using skills, specialized agents, hooks, and tool integrations.
• Understanding of context, execution loops, state, failure recovery, and completion verification.
• Capability to turn an open-ended goal into practical steps, block blockers, and bring others along.
• Capability to take responsibility for the result, including code written by an agent.
• English proficiency at B2 level or higher.
• Nice to have Experience building a coding-agent harness from scratch.
• Familiarity with custom tools or MCP integrations.
• Experience with agent evaluations.
• Experience coordinating multiple agents.

Responsibilities

• Take a goal and make it happen: understand the problem, challenge assumptions, choose an approach, and follow through
• Deliver results that may include a production feature, a validated architecture, or an AI workflow that makes the team more effective
• Put coding agents to work to explore codebases, implement features, test, debug, and review changes
• Decide what to delegate to agents, where to intervene, and how to verify the result
• Configure and extend agent harnesses, including context, skills, specialized agents, hooks, tools, execution environments, and feedback loops
• Connect agents to real repositories and engineering workflows
• Try new approaches on real problems and measure what improves
• Help teammates adopt approaches that work across the team

Tools

Claude Code

Codex

Cursor

Job description

We are looking for a Senior AI-Native Engineer who still gets a kick out of understanding a tricky system, tracking down a stubborn bug, or shipping something people actually use. At EPAM, you'll solve real client problems, make architectural decisions, and help your team find better ways to build software using AI to move faster.

Responsibilities

• Take a goal and make it happen: understand the problem, challenge assumptions, choose an approach, and follow through
• Deliver results that may include a production feature, a validated architecture, or an AI workflow that makes the team more effective
• Put coding agents to work to explore codebases, implement features, test, debug, and review changes
• Decide what to delegate to agents, where to intervene, and how to verify the result
• Configure and extend agent harnesses, including context, skills, specialized agents, hooks, tools, execution environments, and feedback loops
• Connect agents to real repositories and engineering workflows
• Try new approaches on real problems and measure what improves
• Help teammates adopt approaches that work across the team

Requirements

• 5+ years of experience shipping and maintaining production software, with the ability to write code, reason about architecture, debug unfamiliar systems, and explain technical decisions
• Daily experience with coding agents such as Claude Code, Codex, Cursor, or equivalents
• Understanding of agent capabilities, limitations, and failure modes
• Hands-on experience configuring and adapting agent harnesses using skills, specialized agents, hooks, and tool integrations
• Understanding of context, execution loops, state, failure recovery, and completion verification
• Capability to turn an open-ended goal into practical steps, work through blockers, and bring others along
• Capability to take responsibility for the result, including code written by an agent
• English proficiency at B2 level or higher
• Nice to have Experience building a coding-agent harness from scratch
• Familiarity with custom tools or MCP integrations
• Experience with agent evaluations
• Experience coordinating multiple agents