AI Software Engineer

Second Front Systems is hiring an AI Software Engineer to build and deploy production agentic AI systems and the cloud infrastructure that supports them. The role focuses on agent architecture, LLM integration, evaluation, and advising Product and Engineering teams on AI workflows and responsible governance.

Responsibilities:

• Architect production agentic AI systems using LangGraph and LangChain with focus on reliability, traceability, and measurable output quality
• Build agent components including ingestion pipelines, embedding strategies, chunking, retrieval, and re-ranking, and tool suites across managed knowledge-base services, vector databases, and internal and external APIs
• Connect agents to corporate tools and knowledge through Model Context Protocol, constructing both MCP servers and consuming clients
• Establish and operate cloud infrastructure for R&D AI systems: Amazon EKS, Terraform, Helm, and GitOps delivery patterns
• Measure agent and LLM output quality through evaluation and regression suites covering groundedness, faithfulness, relevance, and retrieval quality, running in CI
• Partner with Product teams to scope high-leverage AI opportunities across workflows and advise on agent design, model selection and output quality standards

Requirements:

• 2+ years full-time software, ML, or AI engineering experience with hands-on production work building and shipping LLM-based systems
• Hands-on experience building agentic AI systems with tool-calling agents and multi-step orchestration using LangGraph, LangChain, or equivalent
• Production experience with Retrieval-Augmented Generation including vector stores, embedding models, chunking strategies, and retrieval quality optimization
• Experience deploying containerized AI services to cloud environments using Docker, Kubernetes (EKS preferred), and infrastructure as code (Terraform preferred)
• Demonstrated ability to communicate AI concepts and recommendations to non-technical stakeholders and cross-functional teams
• U.S. citizenship
• Experience building MCP servers or clients, or integrating agents with enterprise tools and knowledge bases
• Experience with AWS AI services such as Bedrock Knowledge Bases, AgentCore, and OpenSearch
• Experience operating LLM gateways or serving infrastructure (e.g., LiteLLM, vLLM, GPU-backed self-hosted models), including single- and multi-tenant designs
• Experience driving AI adoption across a product organization: not just building, but enabling others to use AI effectively
• Experience working in or with the Department of Defense, federal agencies, or regulated environments, and familiarity with compliance frameworks such as NIST 800-53, FedRAMP, or NATO D32