Senior AI Coding Engineer

CGI•Lafayette, LA

2d•Onsite

About The Position

CGI is looking for a Senior AI Coding Engineer to lead the rollout, configuration, and ongoing support of enterprise coding agent tools such as GitHub Copilot and OpenAI Codex. This is a hands-on role for someone who has actually run these platforms in a live production environment — not just built a proof of concept. This position is required in one of the following locations: Birmingham, AL, Knoxville, TN or Lafayette, LA. Your future duties and responsibilities: You'll be the go-to person for getting coding agents working safely and effectively across engineering teams: setting up org level policies, managing access and permissions, tuning agent behavior, and making sure everything holds up from a security standpoint. You'll also build out the supporting tooling — APIs, automation scripts, test harnesses — needed to keep these agents reliable and secure at scale, primarily within an AWS environment. This role calls for someone comfortable working independently, digging into tricky integration issues, and setting the technical direction for how coding agents get adopted across the organization. You'll partner closely with engineering teams to troubleshoot problems, validate agent behavior, and enforce least privilege access for both human users and autonomous agents.

Requirements

5+ years hands-on experience running GitHub Copilot and/or OpenAI Codex in a production setting, including configuring, administering, and troubleshooting.
Experience managing enterprise-level policies: user/group access, repo permissions, tool access, model configuration, and usage controls.
Experience building and securing MCP (Model Context Protocol) servers, including scopes, permissions, auth, identity propagation, credential handling, and least privilege boundaries.
Solid understanding of identity and access management concepts: SSO, SAML, OAuth 2.0, OIDC, SCIM, RBAC, JWT/bearer tokens, and the distinction between authentication and authorization for both people and agents.
Strong AWS background, particularly with Bedrock/Bedrock Agents, IAM, ECS, Lambda, API Gateway, CloudWatch, Secrets Manager, and general VPC/networking knowledge.
Strong Python skills for building REST APIs, automation utilities, integration services, and test/validation frameworks.
Background in security testing for agents, including unauthorized access scenarios, privilege escalation, token handling, prompt injection/tool abuse testing, and building repeatable evaluation frameworks.
Experience building or configuring custom agents, GPTs/assistants, tool integrations, and multi-agent workflows.

Nice To Haves

Bash/shell scripting
PowerShell
Dedicated AI security testing experience

Responsibilities

Lead the rollout, configuration, and ongoing support of enterprise coding agent tools like GitHub Copilot and OpenAI Codex.
Set up organization-level policies for coding agents.
Manage access and permissions for coding agents.
Tune agent behavior and ensure security compliance.
Build supporting tooling, including APIs, automation scripts, and test harnesses, for agent reliability and security at scale within an AWS environment.
Work independently to address integration issues and set technical direction for coding agent adoption.
Partner with engineering teams to troubleshoot problems, validate agent behavior, and enforce least privilege access.