Staff Machine Learning Engineer, Public Sector

Responsibilities

• Lead the architecture and implementation of agentic AI systems focused on long-horizon reasoning, orchestration, and system reliability.
• Build and scale agents for geospatial reasoning over maps and spatial data.
• Design and improve retrieval systems across large collections of static and semi-structured documents.
• Fine-tune and evaluate embedding models for mission-critical datasets.
• Design memory systems for persistent state, long contexts, and learning from prior interactions.
• Own and evolve shared agentic infrastructure and core libraries.
• Define evaluation strategies, robustness tests, failure-mode analyses, and production regression tests for agentic systems.
• Partner with engineering managers, product leaders, and researchers to scope initiatives and unblock execution.
• Mentor engineers and raise standards for system design, ML rigor, and production readiness.
• Travel approximately 10% for customer interaction and team needs.

Requirements

• Active TS security clearance.
• 8+ years building and deploying applied machine learning systems in production.
• Experience with agentic systems, autonomous workflows, or multi-step reasoning and acting systems.
• Strong ML systems engineering background, including model serving, pipelines, monitoring, and evaluation.
• Hands-on experience with retrieval systems, embeddings, or representation learning.
• Proficiency in Python and modern machine learning frameworks such as PyTorch.
• Ability to design systems end to end and operate at Staff-level scope.
• Experience setting technical direction, owning ambiguous problems, and taking initiatives from 0 to 1 through production.
• Ability to balance performance, cost, reliability, and development velocity.

Nice-to-haves

• Experience deploying ML systems in air-gapped, classified, disconnected, on-premises, or customer data-center environments.
• Experience with the Department of Defense, intelligence community, or federal mission users.
• Experience with geospatial data or GEOINT, including maps, imagery, or spatial reference systems.
• Experience with model adaptation, embedding-model fine-tuning, instruction tuning, LoRA/PEFT, or RLHF.
• Experience building evaluation infrastructure for non-deterministic systems, including LLM-as-judge, agent regression suites, or production drift detection.
• Experience turning forward-deployed prototypes into supported and documented capabilities.

Compensation

• Base salary for Washington, DC: $274,400–$343,000 USD.
• Eligible roles may include equity and benefits such as health, dental and vision coverage, retirement benefits, a learning and development stipend, PTO, and potentially a commuter stipend.