Senior AI Engineer

Air•Pittsburgh, PA

About The Position

Air is seeking an experienced Senior AI Engineer to join its Agentic AI team and scale AI capabilities across U.S. government customers. The team is building the systems and infrastructure needed to move its AI agent, Ace, beyond demos and prototypes into reliable, production-grade software for complex real-world problems. This role combines applied AI and systems engineering. The engineer will design and build agent architectures, model integrations, tools, evaluation systems, and infrastructure, working across experimentation and model behavior through production services and distributed infrastructure. The role calls for an engineer who can reason about how agents use context, tools, models, memory, and compute, and turn those concepts into dependable production systems. This is a full-time position based in Air’s Pittsburgh, Pennsylvania office or available remotely. The role may require up to 25% travel, including periodic visits to the Pittsburgh and Arlington, Virginia offices for collaboration, planning, and in-person meetings.

Requirements

U.S. citizenship.
At least 5 years of experience building production software, AI/ML systems, distributed systems, or similar technical systems.
Deep experience designing, building, and operating production AI agents or agent platforms.
Experience developing core agent infrastructure, such as agent runtimes, tool execution, context management, memory, orchestration, or related platform capabilities.
Strong understanding of state-of-the-art agent architectures and the engineering tradeoffs involved in reliable, production-grade agentic systems.
Experience designing agents that use tools, reason across multi-step tasks, interact with external systems, and operate over long-running or complex workflows.
Experience building agent evaluation systems, including task-level and behavioral evaluations, regression testing, and production quality measurement.
Strong understanding of modern LLM systems, including model inference, context engineering, structured outputs, tool calling, retrieval, model selection, and techniques for improving agent performance.
Ability to evaluate emerging models, research, and agent techniques and translate promising approaches into production systems.
Strong Python programming experience and experience building production-quality software.
Experience designing scalable APIs, services, asynchronous systems, and event-driven architectures.
Experience operating production services using Kubernetes and cloud platforms such as AWS, GCP, or Azure.
Strong understanding of distributed systems, containers, service orchestration, networking, storage, and scalable architectures.
Ability to debug complex failures spanning model behavior, agent execution, application code, and production infrastructure.
Ability to move between research and engineering by prototyping ideas, evaluating them rigorously, understanding their outcomes, and turning successful approaches into reliable production systems.
Comfort working in a rapidly evolving field where established patterns may not yet exist.

Nice To Haves

Current possession of a U.S. security clearance, or the ability to obtain one with company sponsorship.
Experience in or exposure to a startup or other entrepreneurial environment.
Experience building secure code execution environments or sandboxes for AI agents.
Experience with multi-agent architectures, agent-to-agent communication, or distributed agent execution.
Experience with fine-tuning, post-training, reinforcement learning, or synthetic data generation.
Experience building AI observability, tracing, and debugging infrastructure.
Experience optimizing inference latency, throughput, GPU utilization, or model-serving costs.
Experience with AI security, adversarial testing, or securing agentic systems.
Experience working in government, defense, or other mission-critical environments.

Responsibilities

Design, build, and improve production agentic AI systems for complex real-world problems.
Develop agent architectures for reasoning, planning, tool use, context management, memory, and multi-step task execution.
Build capabilities that let agents securely interact with data, APIs, code, and external systems.
Develop model and inference infrastructure supporting commercial and open-weight language models.
Evaluate new models, inference techniques, and emerging AI capabilities for potential production use.
Build automated evaluation frameworks to measure agent quality, reliability, task completion, and regressions.
Develop datasets, benchmarks, and evaluation methods for complex agentic workflows.
Improve agent performance through prompt and context engineering, model selection, tool design, inference strategies, and architectural improvements.
Build scalable APIs, services, and infrastructure for agent execution and AI-powered product experiences.
Design systems for asynchronous and long-running agent workflows.
Build infrastructure for safe and reliable execution of agent-generated code and other computational workloads.
Improve observability with structured logging, metrics, distributed tracing, dashboards, and automated alerting.
Investigate failures across models, agents, application code, and distributed infrastructure, and turn findings into systematic improvements.
Optimize model and agent systems for latency, throughput, reliability, and infrastructure cost.
Translate AI research and emerging techniques into practical improvements to production systems.
Work with product, platform, security, and domain teams to take AI capabilities from experimentation to production.