Sr. AI Engineer - IT

LogicMonitor•Austin, TX

6d•$136,500 - $189,420

About The Position

LogicMonitor is seeking a hands-on engineer to implement, improve, and support assigned components of its AI Fabric, an enterprise AI operating layer for building and running AI agents and workflows with trusted context, approved tools, appropriate permissions, and operational controls. The engineer will implement features and services, investigate problems, propose solutions, and make day-to-day engineering decisions within agreed architecture and standards. The environment includes Okta for AI Agents (O4AA), Amazon Bedrock AgentCore, Strands, Model Context Protocol (MCP), agent-to-agent patterns, and workflow orchestration. Prior experience with every product or protocol is not expected; the role requires understanding the concepts, learning unfamiliar systems quickly, and applying them reliably. The role is open to candidates based in or near Austin, Texas. LogicMonitor hires within its Centers of Energy, where teams connect, collaborate, and innovate.

Requirements

At least 5 years of relevant hands-on engineering experience, or an equivalent record of delivering substantial backend, platform, infrastructure, AI, or integration systems.
Strong Python engineering skills and practical experience with APIs, cloud services, containers or serverless runtimes, automated testing, and deployment pipelines.
Experience building or operating production services, including monitoring, troubleshooting, incident follow-up, and maintainability improvements.
Hands-on experience with LLM applications or agents involving tools, retrieval, structured outputs, orchestration, or evaluation; understanding of how context selection, model choice, latency, and cost affect results.
Experience implementing and building infrastructure in AWS AgentCore.
Experience implementing secure integrations using identity, authentication, permissions, secrets management, and audit logging.
Sound engineering judgment about timeouts, retries, idempotency, state, rate limits, and recovery.
Ability to turn a defined platform need into a working implementation plan, explain options and evidence, and recognize when architecture, security, policy, or rollout decisions need review.
Experience writing meaningful tests and using production evidence to improve variable AI behavior.
Clear communication with engineering, security, data, and business partners.
Must be authorized to work in the United States on a full-time, permanent basis without requiring new or initial employer-sponsored work authorization. Candidates with transferable valid U.S. work authorization may be considered case by case.

Nice To Haves

Working ability in JavaScript/TypeScript or SQL.
Experience with Okta for AI Agents (O4AA), Amazon Bedrock AgentCore, Strands, Model Context Protocol (MCP), agent-to-agent patterns, or workflow orchestration.

Responsibilities

Implement and support services for agent execution, context assembly, model routing, tool access, workflow orchestration, and enterprise integrations.
Implement the approved skills-store model, including core, personal, team or department, and organization-wide skills, and help the shared main agent select skills for users and tasks.
Build self-service features for users to create, test, manage, and use personal skills, and implement the review and promotion process for team, department, and organization-wide skills.
Implement skill versioning, selection, testing, auditability, and permission checks; surface conflicts or policy gaps with options and evidence for the AI Fabric lead and relevant approvers.
Build and use agent harnesses within agreed runtime patterns, including configuration, skill and tool loading, execution controls, tracing, tests, and failure handling.
Develop agents and services in code, use workflow tools where appropriate, and integrate APIs, data platforms, tools, and external agents while handling authentication, retries, rate limits, duplicate requests, and recovery.
Apply approved identity and authorization patterns and implement permission-aware context assembly using approved source-authority, freshness, citation, and context-window rules.
Implement approved model routing based on task, policy, quality, latency, and cost, with tests and telemetry for routing decisions.
Add evaluations, logging, alerts, runbooks, and usage and cost measures to components built and supported.
Implement and document components using established platform interfaces and patterns; recommend improvements for consideration by the AI Fabric lead.
Contribute to bounded pilots for background agents using approved task limits, authorization, monitoring, review points, and outcome checks.