e&e is seeking an AI Harness Engineer for an onsite contract opportunity in Harrisburg, PA!
We are seeking an AI Harness Engineer to design, build, and evolve an AI-powered software development lifecycle (SDLC) harness that supports development activities from initial requirements through delivery. This role combines strong software engineering fundamentals with hands-on expertise in LLM-powered applications, AI agents, orchestration, evaluation, and systems integration.
The AI Harness Engineer will build agents, skills, workflows, and guardrails that automate and enhance key stages of the SDLC, including governance, architecture, requirements intake, development, code review, and delivery. The ideal candidate brings a product and developer-experience mindset and understands how to build reliable AI-assisted systems while maintaining appropriate human oversight for consequential decisions.
Key Responsibilities
- Design and build an extensible AI-powered SDLC harness capable of supporting the development lifecycle from requirements intake through pipeline-ready delivery.
- Develop AI agents and reusable skills that support governance, architecture, requirements, development, testing, review, and delivery activities.
- Create orchestration workflows that coordinate multiple agents, tools, skills, and lifecycle stages.
- Encode organizational standards, approval processes, compliance requirements, and data-privacy controls as enforceable checks within the harness.
- Support architecture and system design activities while helping ensure implementations remain aligned with approved designs and technical standards.
- Transform requests and Jira work items into structured, actionable, and testable technical specifications.
- Enable agents to assist with software development activities including coding, testing, refactoring, and application scaffolding.
- Build automated code-review and merge-review capabilities to identify issues before human review.
- Integrate the harness with GitLab, Jira, model providers, and other development systems through APIs.
- Provide AI agents with appropriate contextual information, including source code, engineering standards, documentation, and previous architectural or technical decisions.
- Develop evaluations, automated tests, monitoring, and guardrails to measure and improve the reliability of non-deterministic AI systems.
- Design clear human-in-the-loop workflows, approvals, and handoffs to maintain human control over consequential actions.
- Deliver reviewed, pipeline-ready software changes that integrate cleanly with existing CI/CD processes.
- Design the platform for extensibility so additional agents, skills, tools, and SDLC stages can be incorporated without significant rework.
- Maintain the harness as an internal product through versioning, documentation, testing, feedback, and continuous improvement.
- Partner with engineering, architecture, governance, security, and other technical stakeholders to drive adoption and ensure the solution meets organizational standards.
Required Qualifications
- Strong professional software engineering background with demonstrated ability to design, develop, test, and maintain production-quality applications using modern programming languages.
- Hands-on experience developing LLM-powered applications, AI agents, or agentic workflows.
- Experience with agent orchestration, tool and skill design, structured outputs, retrieval, and context management.
- Strong understanding of evaluation methodologies, testing strategies, and guardrails for non-deterministic AI systems.
- Experience integrating applications and platforms through REST APIs or similar interfaces, including development platforms, issue-tracking systems, and model-provider APIs.
- Strong understanding of the full software development lifecycle, including governance, architecture, requirements, development, testing, code review, and delivery.
- Experience working with source control and modern software development practices.
- Ability to translate complex development processes and organizational standards into automated workflows and controls.
- Strong focus on software quality, maintainability, reliability, and security.
- Product and developer-experience mindset with the ability to create tools that improve engineering productivity and adoption.
- Strong communication and collaboration skills with the ability to work across technical and business teams.
Preferred Qualifications
- Experience with Anthropic APIs, Claude, Claude Code, agent skills, or comparable AI agent frameworks and technologies.
- Familiarity with GitLab and GitLab CI/CD, particularly integrating applications and automated workflows with existing pipelines.
- Advanced prompt engineering and context engineering experience.
- Experience delivering agentic or generative AI capabilities to production users.
- Familiarity with MLOps, workflow orchestration, or model-serving technologies such as Airflow, MLflow, or vLLM.
- Experience developing solutions within regulated or data-privacy-sensitive environments, including familiarity with FERPA or comparable regulatory requirements.
- Familiarity with on-premises or self-hosted technology environments.
- Understanding of Kubernetes-based application and infrastructure environments.