Senior AI / Backend Engineer

Us

At Aretec, Inc., we are catalysts for change within the federal government landscape. Specializing in advanced analytics, machine learning, data analysis, cybersecurity, and business optimization, we empower federal agencies to achieve their most critical missions. As a premier partner and prime vendor, we deliver innovative, high-impact solutions that address complex challenges and drive national progress. Our commitment to excellence and innovation positions us at the forefront of transforming governmental operations, enhancing efficiency, and making a lasting difference in the lives of citizens.

You

You are a thoughtful, collaborative technology professional who turns complex challenges into practical, secure, and maintainable solutions. You work comfortably with engineers, product stakeholders, and federal partners; communicate clearly; and balance delivery speed with quality, traceability, and responsible use of AI. You are energized by mission-driven work and eager to contribute within an established team that values sound judgment and continuous improvement.

What We are Looking For

Aretec is seeking a Senior AI / Backend Engineer to build the intelligent services behind a greenfield AI solution. You will develop Python services, LLM integrations, retrieval pipelines, agent workflows, APIs, guardrails, and automated evaluations that produce useful, well-grounded results with clear references to supporting source material.

Key Responsibilities

• Design and develop Python backend services and APIs for the greenfield AI solution.
• Integrate large language models and build retrieval-augmented generation pipelines over approved enterprise data sources.
• Implement agent orchestration and reusable prompt-framework components.
• Build retrieval, ranking, context-assembly, and citation mechanisms that tie summaries and findings to supporting documents.
• Develop guardrails that constrain unsupported assumptions, overly prescriptive outputs, legal determinations, and decisions that belong to human staff.
• Create deterministic and probabilistic evaluation automation that measures accuracy, completeness, consistency, grounding, and adherence to guardrails.
• Build automated unit, integration, regression, and AI-quality tests.
• Instrument services for troubleshooting, performance analysis, and continuous improvement.
• Work closely with the technical lead, data engineer, full-stack engineer, and enterprise-search stakeholders throughout design and delivery.

The Skills We are Looking For

Required Qualifications

• Strong Python software-development skills.
• Practical experience integrating LLMs into production or production-oriented applications.
• Experience designing RAG systems, retrieval pipelines, embeddings, context construction, or comparable information-retrieval workflows.
• Experience with RESTful APIs and backend service design.
• Understanding of agent orchestration, prompt design, AI guardrails, and model-output evaluation.
• Experience with automated testing and disciplined software-engineering practices.
• Strong analytical, debugging, documentation, and communication skills.

Preferred Skills

• Experience with source citation, document intelligence, enterprise search, or knowledge-retrieval systems.
• Experience with structured human feedback loops and AI evaluation frameworks.
• Experience delivering applications in federal, regulated, or security-conscious environments.
• Familiarity with identity, access control, observability, and secure API integration.

The Expectations of the Job

Day One

• Learn the greenfield AI solution use cases, enterprise-search environment, available data sources, security boundaries, and current AI components.
• Align with the technical lead on development standards and priority workflows.

Day Thirty

• Deliver initial Python/API contributions and prototype or improve a priority retrieval or LLM workflow.
• Add foundational tests and quality measurements for early AI capabilities.

Day Sixty

• Own significant RAG, agent, guardrail, or evaluation components and integrate them with the application.
• Use system tests and user feedback to improve grounding, accuracy, completeness, and appropriate scope.

Day Ninety

• Take end-to-end ownership of core AI/backend capabilities from design through production-ready validation.
• Help establish repeatable evaluation, monitoring, and improvement practices for the gr…