EPAM

ai engineer for coding agents

In short

This lead-level remote role at EPAM focuses on using and configuring coding agents (Claude Code, Codex, Cursor, or equivalents) to explore codebases, implement features, and review agent-written changes. It requires production software experience, strong architecture and debugging skills, and daily hands-on work with agent harnesses, hooks, tools, and execution loops. Candidates should be able to turn open-ended goals into practical steps and take responsibility for outcomes. Experience building custom agent tooling, MCP integrations, or agent evaluations is a plus.

Description
EPAM helps clients solve engineering problems by building production software and applying AI to improve software development.
Responsibilities
•Turn goals into results by understanding problems, challenging assumptions, choosing approaches, and following through
•Use coding agents to explore codebases, implement features, test, debug, and review changes; decide what to delegate and how to verify the results
•Configure and extend agent harnesses, including context, skills, specialized agents, hooks, tools, execution environments, and feedback loops
•Connect agents to repositories and engineering workflows
•Test approaches on real problems, measure their impact, and help teammates adopt effective methods
Requirements
•Have shipped and maintained production software
•Be able to write code, reason about architecture, debug unfamiliar systems, and explain technical decisions
•Have daily experience with coding agents such as Claude Code, Codex, Cursor, or equivalents, including understanding their capabilities, limitations, and failure modes
•Have hands-on experience configuring and adapting agent systems using skills, specialized agents, hooks, and tool integrations
•Understand context, execution loops, state, failure recovery, and completion verification
•Turn open-ended goals into practical steps, work through blockers, bring others along, and take responsibility for outcomes, including agent-written code
Nice to have
•Building a coding-agent harness from scratch
•Custom tools or MCP integrations
•Agent evaluations
•Coordinating multiple agents
Conditions
No conditions specified in the posting.

Stop applying.
Get referred.

$10 in free credits

In just 60 seconds, fully on autopilot — Eraldt's AI agents read your CV, map the roles you fit, find live vacancies and the people who can refer you, write personal emails, and get you referred.

This isn't an application: your CV doesn't go to the company, and we don't promise a referral for this exact role.

Discover who else lands interviews on autopilot