Government and Infrastructure - AI and Data - AI Automation Senior Engineer
Ernst and Young1 day ago
Base Salary
$107k - $201k/yr
Responsibilities
- Build and operate automated ingestion pipelines for source code, database schemas and scripts, deployment configuration, and technical documentation.
- Develop static-analysis and LLM-assisted extraction for application components, technologies, dependencies, interfaces, data-access patterns, risks, and technical debt.
- Build LLM capabilities for code and business-logic summarization, capability tagging, similarity analysis, duplicate detection, and modernization recommendations.
- Implement RAG and hybrid keyword/vector search with source citations over portfolio evidence.
- Load extracted entities and relationships into graph and relational stores for dependency, impact, clustering, and modernization analysis.
- Create evaluation harnesses, golden datasets, grounding and faithfulness checks, regression tests, and quality metrics for prompts and agents.
- Design human-in-the-loop review and traceability processes covering reviewer decisions, model versions, prompt versions, and rationale.
- Instrument and optimize pipelines for observability, throughput, performance, and token or cost consumption.
- Package prompts, agents, tools, and pipeline components as reusable documented assets.
- Own AI capabilities or pipeline components end to end from requirements through production operation.
- Design multi-step agent workflows and tool integrations, including Model Context Protocol tools, guardrails, and failure handling.
- Select and tune models, embeddings, chunking, and retrieval approaches using evaluation results.
- Review Staff engineer code, prompts, and evaluation results and coach engineers on AI engineering practices.
- Explain AI-derived findings, methods, and limitations to client technical stakeholders and application owners.
Requirements
- Bachelor’s degree in computer science, software engineering, information systems, computer engineering, or a related field, or equivalent practical experience.
- 3–6+ years of software, data, or ML engineering experience, including production AI/ML system architecture and delivery.
- Experience leading teams of five or more engineers and delivering to clients in federal or regulated environments.
- Eligibility to obtain and maintain a Secret Clearance.
- Experience integrating generative AI models from OpenAI, Anthropic, or open-source LLM providers into web-based products.
- Experience working with designers, product managers, and AI/ML engineers to create intelligent user experiences.
- Experience writing tests and participating in code reviews.
- Preferred experience with static code analysis and parsing, graph databases and algorithms, agent frameworks, LLM evaluation and observability tooling, Azure services, Azure Government, legacy application modernization, federal delivery, and Azure AI Engineer Associate certification or comparable certification.
Benefits
- Medical and dental coverage, pension, and 401(k) plans.
- Paid time off under a flexible vacation policy, EY holidays, winter and summer breaks, personal and family care leave, and other leaves of absence.
- Work may occur at client, EY, or contractor sites, with in-person work as needed in the Washington, DC area.
- Travel of 20–30% or more may be required, and assignments are generally within a commutable distance of the work-location office.
- Base salary varies by location from $106,900 to $200,600.
Tech Stack
Categories
AI ApplicationsData Engineering
About Ernst and Young
Ernst & Young (EY) provides audit/assurance, tax, consulting, strategy and transactions services to enterprises, financial institutions, and public‑sector clients. Structured as a global network of partner‑owned member firms, it sells professional services on a fee basis, including a dedicated Financial Services Organization for banking, insurance, and capital markets. Headquartered in London, EY was formed in 1989 from the merger of Ernst & Whinney and Arthur Young, and operates in 150+ countries.