Lead AI Developer

EPAM Systems Inc; United States; Remote; USD 140,000 - 190,000; Full time

Job summary

EPAM Systems is seeking a hands-on Lead AI Developer to partner with our Lead AI Engineer. You will write production-grade C#/.NET code, build stateful agentic workflows, and deploy automated pipelines that accelerate SDLC across multiple teams.

You will integrate with Microsoft and GitHub tooling, mentor engineers through pair programming, and help optimize token usage and model routing while enforcing guardrails and code quality for scalable AI solutions.

Qualifications

• Minimum 5 years of hands-on experience.
• At least 1 year of leading and managing teams.
• Deep coding proficiency in C# and .NET Core, including architecture and backend tooling.
• Experience building production-grade LLM apps and stateful agent execution loops.
• Experience automating software pipelines with GitHub Actions and Copilot.

Responsibilities

• Construct and roll out stateful multi-agent workflows and LLM parsers in C#/.NET.
• Convert specs and JSON schemas into machine-readable formats for autonomous code generation.
• Build extensions and scripts that plug into GitHub Copilot, GitHub Actions, and Microsoft toolchains.
• Coach engineers through pair programming and hands-on mentoring to drive adoption.
• Fine-tune token usage and model routing to reduce costs and improve performance.
• Apply engineering guardrails and code-quality parsers to ensure safe AI code.

Skills

C#/.NET

Team leadership

CI/CD pipelines

GitHub Copilot

Tools

GitHub Actions

GitHub Copilot

Cloud-native design

Job description

We are looking for a highly hands‑on, execution‑focused Lead AI Developer to serve as the ground‑level technical partner to our Lead AI Engineer. While the Lead AI Engineer establishes the architectural vision and transformation roadmap, you will be the core builder in the trenches—writing production‑grade C#/.NET code, engineering stateful agentic workflows, and implementing automated pipelines. Your primary mission is to revolutionize our Software Development Life Cycle (SDLC) and supercharge developer productivity by building, tuning, and deploying agentic AI tools that seamlessly integrate into the client's modern Microsoft and GitHub engineering ecosystem across multiple teams.

Responsibilities

• Construct, validate, and roll out stateful multi‑agent workflows, tool/function‑calling mechanisms, and structured LLM output parsers in C#/.NET, automating SDLC phases such as requirements translation, code generation, and test suite creation.
• Convert specifications, JSON schemas, and structured technical documentation into machine‑readable formats that power autonomous multi‑agent code and test generation spanning multiple repositories.
• Build custom extensions, plugins, and automation scripts that plug into GitHub Copilot, GitHub Actions, and Microsoft developer toolchains to simplify workflows and reduce mental overhead for developers.
• Embed directly with engineering teams through pair programming, live troubleshooting, and hands‑on coaching to model effective daily adoption of agentic AI tooling.
• Fine‑tune token optimization and FinOps practices, covering context‑window management, prompt caching, and hybrid routing between Small Language Models and frontier models to reduce operational token spend.
• Apply engineering guardrails, security checks, and code‑quality parsers established by the Lead AI Engineer to keep AI‑generated code safe, compliant, and production‑ready.
• Collect day‑to‑day, tactical input from developers on workflow friction points and relay findings to the Lead AI Engineer to continually sharpen prompt sets, agent loops, and overall developer experience.

Requirements

• A minimum of 5 years of relevant hands‑on experience.
• At least one year of experience in leading and managing teams.
• Deep, current coding proficiency in modern C# and .NET Core, including solid understanding of architecture, cloud‑native design patterns, and building resilient backend tooling.
• Demonstrated hands‑on background building production‑grade LLM applications, stateful agent execution loops, and structured outputs such as JSON mode/function calling using GitHub Copilot.
• Real‑world experience automating software engineering pipelines, custom extensions, and CI/CD workflows through GitHub Actions, GitHub Copilot, and modern developer tooling.
• Working knowledge of spec‑driven development, including using structured schemas such as JSON Schema to guide automated code generation and multi‑agent workflows.
• Sharp awareness of developer friction points throughout the SDLC, paired with genuine enthusiasm for building tools that boost developer flow, satisfaction, and speed.
• Action‑driven, hands‑on working style, ready to partner closely with engineering teams to coach them through practical agentic AI transformations.

Nice to have

• Advanced GitHub Copilot proficiency backed by GH-600/300 certification.