Senior / Staff Agent Platform Engineer

We are hiring a Senior or Staff Agent Platform Engineer to build the agent and workflow systems that connect customer data, a structured knowledge base, organizational automation, and a geospatial operating picture. The role covers multi-step agent workflows, retrieval, tool design, durable execution, human approval, and permission-aware access to customer data.

This is a hands-on individual contributor role on an existing product and engineering team. You will write production code regularly and develop shared capabilities across the agent runtime, workflow, retrieval, and model-serving layers. The position is on-site and requires U.S. citizenship and eligibility to work on ITAR/export-controlled programs.

• Design and implement agent workflows for multi-step research, analysis, and task execution over structured and unstructured customer data.
• Build reliable tool interfaces for retrieval, data analysis, geospatial operations, and organizational workflows.
• Develop durable execution patterns for long-running work, including checkpoints, resumable runs, retries, partial failure, cancellation, and recovery.
• Implement human-in-the-loop controls for consequential actions, including previews, approvals, structured review, and clear failure states.
• Improve retrieval across semantic search, structured datasets, and knowledge graphs, including query planning, evidence assembly, and citations.
• Ensure agents operate with the same tenant, role, resource, and data-access permissions as the person they represent.
• Carry authorization and audit requirements through retrieval, tool execution, delegated work, generated artifacts, and exports.
• Optimize agent execution for a range of deployment targets, including analyst laptops, local GPU clusters, and cloud-connected environments.
• Profile and improve end-to-end latency, throughput, context use, memory consumption, and inference cost across model, retrieval, and tool-execution paths.
• Fine-tune and evaluate open-source models for targeted agent capabilities such as tool use, structured extraction, retrieval, planning, and domain-specific reasoning.
• Develop model-selection, routing, quantization, and fallback strategies that preserve useful agent behavior across different compute and connectivity constraints.
• Build evaluation systems for task completion, groundedness, retrieval quality, tool reliability, latency, and permission enforcement.
• Add production observability that distinguishes model, retrieval, orchestration, and tool failures.
• Define stable platform interfaces that allow agent, workflow, retrieval, and product capabilities to evolve without duplicating business logic.
• Collaborate with product, frontend, data, ML, security, infrastructure, and customer-facing engineers.
• At the Staff level, set technical direction for agent architecture, establish engineering standards, and mentor other engineers.

• U.S. citizenship, with eligibility to work under ITAR/export-control programs. Active or prior security clearance is a plus.
• Bachelor's or Master's degree in Computer Science or a related technical field, or equivalent practical experience.
• 5+ years building and operating production software systems; Staff-level candidates will typically bring broader technical leadership experience.
• Strong Python engineering skills and experience with distributed backend systems, APIs, databases, and asynchronous or event-driven execution.
• Experience shipping an agent, retrieval, search, automation, or workflow product used by real users.
• Practical experience with tool calling, context management, structured outputs, failure recovery, and long-running workflows.
• Experience with semantic retrieval and at least one structured retrieval approach such as SQL, graph traversal, or analytical query systems.
• Experience training, fine-tuning, adapting, or evaluating open-source language models using methods appropriate to the use case, such as supervised fine-tuning, parameter-efficient fine-tuning, distillation, or preference optimization.
• Experience deploying or optimizing inference across at least two compute environments, such as local workstations, GPU clusters, edge devices, or cloud infrastructure.
• Experience translating product requirements into evaluation datasets, measurable outcomes, and regression tests.
• Sound judgment about when to use model autonomy, deterministic controls, policy enforcement, or human approval.
• Clear, concise written and verbal communication.

• Experience with multi-tenant authorization, RBAC, ABAC, or permission-aware retrieval.
• Experience with Temporal or another durable workflow and orchestration system.
• Experience with knowledge graphs, hybrid retrieval, entity-centric search, or data provenance systems.
• Experience with geospatial analysis, mapping applications, or spatial data systems.
• Familiarity with MCP, agent-to-agent protocols, or equivalent tool and service orchestration patterns.
• Experience deploying software on-premises, at the edge, or in air-gapped environments.
• Experience with local model-serving frameworks and production inference systems, including quantization, batching, caching, and hardware-aware model selection.
• Familiarity with CUDA, GPU performance profiling, distributed inference, or cluster scheduling.
• TypeScript or React experience sufficient to collaborate effectively on product interfaces.
• Experience in defense, intelligence, public safety, or another high-consequence domain.