Data and Automation Engineer (Federal Grants AI Platform)

Location: Remote (U.S.), with occasional on-site meetings in the Washington, DC metropolitan area

Employment type: Full-time

Eligibility: U.S. citizenship required; must be able to obtain and maintain a Federal Public Trust background investigation

About the role

Zenius is hiring a data and automation engineer to build and run the data and workflow layer of an AI-assisted audit resolution platform for a Federal grantmaking agency. The platform supports a team of grants and audit specialists working through a large volume of single audit and Office of Inspector General audit findings on Federal grant awards. You will bring audit data in from public Federal sources and the agency's own systems, turn audit report packages into structured records that people can review, and build the queues, logs, and dashboards that show every finding and every action taken on it. This is production engineering on real Federal records in an authorized cloud environment. Every output must be traceable: where the data came from, which model or rule produced the result, how confident it was, and which person accepted or changed it. Engineers who build logging, tests, and evaluation in from the first commit will do well here. The platform supports decisions; agency officials retain all decision-making authority.

Responsibilities

• Build and operate data pipelines that retrieve audit records from the Federal Audit Clearinghouse public data interface, ingest extracts from the agency's audit tracking system through read-only access, and take in Inspector General audit reports.
• Reconcile every load so that what was retrieved equals what was stored, and surface exceptions for the analyst team to clear.
• Implement the finding record as an append-only ledger: one entry per finding, with every machine and human action added to it and nothing overwritten.
• Build document extraction for single audit packages that captures the finding number, program and Assistance Listing number, compliance requirement, condition, cause, effect, questioned costs, repeat-finding status, corrective action plan, and prior-year status.
• Implement decision-support components on the platform's agent layer so that every output carries a source citation, a confidence score, and the name of the reviewer who accepted or changed it.
• Build the workflow layer: work queues, review and approval steps, audit logging, and dashboards the client can open at any time.
• Instrument drift and quality monitoring, set confidence thresholds and re-routing rules, and take any component that falls below threshold out of service until it is retested.
• Build and maintain held-out evaluation sets drawn from real audit reports, and run the evaluations each capability must pass before it is used on live work.
• Implement the security controls that apply to application development, logging, and configuration management, and supply the evidence needed for the system's security authorization.
• Produce the technical documentation, configurations, prompts, workflow definitions, and evaluation sets that are delivered to the client under the contract.
• Deliver in short increments alongside the Intelligent Automation Lead, the security specialist, and the audit resolution team, demonstrating working software and acting on feedback from the people who use it.

Requirements

• A bachelor's degree in computer science, engineering, information systems, or a related field and one to two years of relevant experience, or four or more years of relevant experience in place of a degree.