Senior Agentic AI Engineer

We respectfully request that 3rd parties refrain from contacting us regarding this posting.

Overview

LABUR is partnering with a client to find a Senior Agentic AI Engineer who will design, build, and productionize agentic AI systems that automate research, data collection, enrichment, validation, and decision-support workflows. This role centers on delivering reliable, scalable AI solutions using LLMs, retrieval systems, orchestration frameworks, and human-in-the-loop controls that drive measurable business outcomes. The ideal candidate combines strong engineering fundamentals with practical judgment around when agentic patterns are appropriate, with a focus on reliability, evaluation, traceability, and long-term maintainability. This role is fully remote.

Responsibilities

• Design and develop agentic AI systems that automate research, enrichment, validation, and decision-support processes using LLMs, retrieval systems, structured data, external tools, and MCP services
• Implement orchestration logic, routing, state management, retries, checkpointing, fallback strategies, and failure recovery mechanisms
• Architect and maintain RAG, search, and data-ingestion pipelines across structured and unstructured data sources
• Establish evaluation frameworks, testing strategies, observability, and quality controls for AI agents and workflows
• Implement provenance, traceability, governance, and human-in-the-loop review processes where appropriate
• Collaborate with product, data, platform, and engineering teams to identify high-value automation opportunities
• Support production operations, incident response, root-cause analysis, and continuous improvement of deployed AI systems

Qualifications

• Strong Python and production software engineering skills with hands-on experience building and operating production agentic AI systems
• Experience with orchestration frameworks such as LangGraph, Airflow, AWS Step Functions, or Temporal, including tool use, function calling, MCP integration, and agent harness engineering
• Proficiency designing state management, retries, checkpointing, routing, fallbacks, and failure recovery patterns
• Deep understanding of retrieval, RAG, search, and data-ingestion pipelines across structured and unstructured data sources
• Experience with AI evaluation, testing, observability, quality assurance, and human-in-the-loop workflows with provenance and data-quality controls
• AWS experience required; Snowflake experience preferred
• Strong systems-design skills and architectural judgment, including familiarity with enterprise AI governance, security, and compliance

Compensation

$80-$85/hr - Dependent on fit and experience