Mid-level Applied Machine Learning Engineer – Agentic Workflows

Athenaworks is seeking a Mid-level Applied Machine Learning Engineer to build advanced AI capabilities for litigation, investigations, and data breach workflows involving large-scale document collections. The role focuses on developing LLM, RAG, and agentic decision-support systems that provide grounded, explainable, auditable, and reliable outputs for legal and breach-related review decisions.

Responsibilities:

• Design and implement applied ML and GenAI solutions that support document review and analysis across litigation, investigations, and data breach workflows
• Build LLM-powered systems that enable corpus-level reasoning, allowing users to ask questions of large document sets and receive grounded, evidence-backed outputs
• Design and deploy retrieval-augmented generation (RAG) approaches that combine semantic retrieval, document context, and generative reasoning to produce explainable results
• Develop agentic or multi-step reasoning workflows that decompose complex review for analysis tasks into bounded, auditable decision steps
• Build systems that generate summaries, classifications, and decision rationales relevant to review quality, responsiveness, privilege, confidentiality, and exposure assessment
• Apply grounding and citation techniques to ensure model outputs remain anchored in authoritative document content and supporting evidence
• Design decision-support logic that incorporates confidence thresholds, review gating, and human judgment where appropriate
• Integrate AI services into production-grade systems with attention to performance, scalability, auditability, and failure handling
• Collaborate with product, platform, and domain experts to ensure solutions align with real-world litigation and breach response practices
• Monitor system behavior in production and continuously improve outcomes based on user feedback and observed decision quality

Requirements:

• Clearance: U.S. Citizen, Public Trust eligible
• Hands-on experience working in teams utilizing large language models (LLMs), retrieval-augmented generation (RAG), embeddings, and agentic AI techniques in user-facing applications
• Strong background in applied machine learning with experience deploying ML systems in production environments
• Experience designing or operating semantic retrieval, ranking, or AI-augmented search systems over large document collections
• Experience building systems that operate over large, heterogeneous, and imperfect document datasets
• Ability to design grounded, explainable AI outputs that support defensible legal and breach-related decision-making
• Experience translating ambiguous legal, investigative, or compliance requirements into measurable technical behavior
• Strong software engineering fundamentals, including testing, versioned deployments, and maintainability
• Comfort operating in domains where accuracy, consistency, automation, and risk must be carefully balanced
• Experience with eDiscovery, litigation support, internal investigations, or data breach response
• Familiarity with legal review concepts such as responsiveness, privilege, confidentiality, issue tagging, and exposure assessment
• Experience working in regulated or compliance-driven environments
• Experience with hybrid AI systems combining learned models with structured signals, rules, or human feedback