Senior AI Engineer – Agentic AI Platform

Hudson Manpower is hiring a Senior AI Engineer to design and build enterprise agentic AI platforms across the full lifecycle of autonomous agents.

Responsibilities

• Design and develop multi-agent AI systems for enterprise business use cases.
• Build autonomous and semi-autonomous AI workflows with orchestration and choreography patterns.
• Implement agent workflow strategies including Supervisor-Worker, Sequential, ReAct, Planner-Executor, and Writer-Critic.
• Create scalable frameworks for agent communication and execution.
• Deliver closed-loop workflows using validation, retry, evaluation, and feedback mechanisms.
• Develop reusable platform capabilities that multiple business teams can leverage.
• Design enterprise AI governance and operational controls.
• Build API-driven AI services that include rate limiting, quota management, authentication, authorization, audit logging, multi-tenant usage tracking, and cost attribution.
• Establish agent onboarding and lifecycle management capabilities.
• Design agent communication using direct calls, event-driven architectures, message queues, and publish-subscribe patterns.
• Work on event-driven workflows using Kafka, Azure Service Bus, and Azure Durable Functions.
• Design AI memory systems across short-term and long-term needs (vector databases, semantic caching, conversation memory, agent state persistence, and RAG).
• Build knowledge orchestration frameworks for agent collaboration.
• Support enterprise knowledge models using graph databases and ontology-driven approaches.
• Create knowledge graphs to enable relationship-based reasoning across structured, unstructured, and graph-based knowledge sources.
• Implement AI consumption governance across business domains, including token and model consumption tracking, API utilization, and operational cost visibility.
• Develop chargeback/showback mechanisms and support FinOps reporting and capacity planning.
• Build observability frameworks for AI workloads, including monitoring for execution, tool usage, latency, hallucinations, failure rates, and model quality.
• Create dashboards and operational metrics for agentic AI systems.
• Implement guardrails and safety controls such as prompt protection, data masking, PII protection, and human-in-the-loop validation.
• Ensure secure agentic AI systems for sensitive business data with enterprise security and governance alignment.
• Develop agent and tool evaluation frameworks; measure response quality and detect hallucinations.
• Implement closed-loop evaluation mechanisms and apply context engineering, prompt engineering, retrieval optimization, agent tuning, and AI benchmarking.

Requirements

• 8–10 years of software engineering or platform engineering experience.
• 3+ years of hands-on AI/ML or Generative AI experience.
• Production experience building enterprise-scale AI applications.
• Strong experience designing AI architectures and platforms, not only individual AI applications.
• Hands-on experience with Agentic AI / AI Agents.
• Strong experience with Azure.
• Hands-on experience with Azure AI Foundry and Azure OpenAI.
• Strong experience with LangChain and/or LangGraph.
• Strong Python development experience.
• Experience with multi-agent orchestration and agentic workflow patterns.
• Experience with RAG, vector databases, AI memory, and agent state management.
• Experience with REST APIs and API gateways, preferably Azure API Management (APIM).
• Experience with event-driven architectures and messaging systems.
• Experience with AI monitoring, observability, governance, and cost/token usage tracking.
• Experience with enterprise data/storage technologies such as Cosmos DB, PostgreSQL, MongoDB, or vector databases.
• Experience with SQL.
• Experience designing scalable, secure, and governed AI platforms.

Technologies

• Azure, Azure AI Foundry, Azure OpenAI
• LangChain, LangGraph
• Python
• Vector databases, RAG
• API gateways, Azure API Management (APIM)
• Kafka, Azure Service Bus, Azure Durable Functions
• REST APIs
• Cosmos DB, PostgreSQL, MongoDB, SQL
• Message queues, publish-subscribe patterns, event-driven architectures

Desirable skills

• Semantic Kernel
• Model Context Protocol (MCP)
• C# / .NET
• Azure Event Grid
• Neo4j, Stardog, Amazon Neptune, or other graph databases
• Enterprise knowledge graphs and ontology-driven AI solutions
• AI FinOps and chargeback/showback
• Responsible AI frameworks
• AI evaluation and benchmarking
• AWS or GCP
• Experience in healthcare, financial services, insurance, or other regulated industries

Location and compensation

• Cincinnati, OH (onsite)
• USD 55–60 per hour