Agent Platform Engineer

Hitachi•Dallas, TX

About The Position

The Agent Platform Engineer designs, secures, and operates AI agents and the tools they invoke, with tracing, evaluation, and cost-management capabilities suitable for production environments. The role develops AI agents and strongly typed tools that integrate with enterprise systems; implements entitlements, agent identities, guardrails, and approval workflows for high-impact actions; and establishes tracing, evaluation, and cost-per-decision monitoring. It manages the full agent lifecycle, including versioning, deployment, and operational monitoring, integrates agents with existing engineering platforms and streaming data sources, and provides operational runbooks for support teams. The role requires Python development and API engineering, experience with LLM application and agent frameworks, and familiarity with observability, evaluation, ontology, knowledge graph, cloud, CI/CD, infrastructure-as-code, identity management, and AI guardrail practices. The company describes itself as an inclusive, equal opportunity employer and notes that flexible arrangements depend on role and location.

Requirements

10+ years of experience in software engineering or platform engineering.
2+ years of experience developing and deploying production-grade LLM or agent-based applications.
Experience designing human-in-the-loop approval workflows and using evaluation frameworks as release gates.
Python development and API engineering, including REST and typed schemas.
Experience with LLM application and agent frameworks such as LangChain, LangGraph, or equivalent, including tool/function calling and Model Context Protocol (MCP).
Experience with LLM evaluation, tracing, and observability tools such as MLflow, LangSmith, OpenTelemetry, or equivalent.
Knowledge of ontologies and knowledge graph technologies such as RDF/OWL, property graph platforms such as Neo4j, or equivalent.
Knowledge of cloud infrastructure, CI/CD pipelines, and infrastructure-as-code practices supporting Agentic Platforms.
Knowledge of identity management, access controls, and guardrail implementation for AI systems.

Nice To Haves

Experience with Databricks (Genie, Mosaic AI, Agent Framework).
Familiarity with AI-assisted software development lifecycle practices.
Automotive or manufacturing industry experience.

Responsibilities

Develop AI agents and strongly typed tools that integrate with enterprise systems.
Implement entitlements, agent identities, guardrails, and approval workflows for high-impact actions.
Establish tracing, evaluation, and cost-per-decision monitoring capabilities (FinOps for AI).
Manage the full agent lifecycle, including versioning, deployment, and operational monitoring.
Integrate agents with existing engineering platforms and streaming data sources.
Provide operational runbooks for support teams.