Agentic AI Engineer - Cloud Infrastructure Automation
Agentic Ai
Agentic Ai Engineer
Agentic Ai Governance
Agentic Ai Security
Agentic Ai Workflows
Ai Agent
Ai Ml
Ai Security
Artificial Intelligence
Artificial Intelligence Engineer
Automation
Cloud
Cloud Infrastructure
Cloud Infrastructure Automation
DevOps
Engineer
Engineering
Generative AI
Generative Ai Agents
Generative Ai Engineering
Infrastructure
Infrastructure As Code
Policy As Code
Security As Code
Job Description
Robotics Technologies is seeking an Agentic AI Engineer to help design and deliver AI agents that can operate directly against live cloud infrastructure. This contract role is centered on governed automation, where agent actions are interpreted from infrastructure state and telemetry, validated with policy controls, and executed with appropriate approval workflows inside the engagement timeframe.
What you’ll do
- Build and deploy agentic workflows using an LLM orchestration framework such as LangGraph, CrewAI, AutoGen, Semantic Kernel, or cloud-native options including Bedrock Agents and Azure AI Agent Service.
- Create tool and function-calling layers that let agents safely interact with infrastructure operations, including MCP-based tooling where applicable.
- Programmatically generate, review, and validate Terraform modules, integrating policy-as-code checks with OPA/Sentinel plus tflint and checkov into the agent’s output paths.
- Connect agent operations to observability and ITSM systems for alert triage and automated ticket-to-action flows across DataDog, ServiceNow, PagerDuty, and Jira.
- Implement guardrails including least-privilege execution roles, approval gates, dry-run or plan-only modes, and audit logging for every agent action.
- Develop evaluation and regression harnesses to track agent behavior such as accuracy, tool-selection correctness, and failure containment.
- Integrate agent workflows into existing CI/CD pipelines.
- Document the architecture and support a handover to the client’s platform team at the end of the engagement.
Role requirements
- 2+ years building production LLM applications, including at least one deployed multi-step agentic system.
- Strong command of one orchestration framework (LangGraph/LangChain, CrewAI, AutoGen, Semantic Kernel) including tool/function calling and structured output.
- Practical experience with agent guardrails, retries, failure handling, and human-in-the-loop approval design.
- RAG fundamentals including embeddings, vector stores (pgvector, OpenSearch, Pinecone), and retrieval quality tuning.
- Agent evaluation and observability with LangSmith, Langfuse, Ragas, or equivalent custom harnesses.
- 5+ years in cloud infrastructure, platform, or DevOps engineering.
- Deep Terraform experience including module design, state and remote backends, workspaces, and drift handling; Terragrunt or Terraform Cloud/Enterprise is a plus.
- Policy-as-code with OPA/Rego, Sentinel, or equivalent.
- Cloud depth in at least one of AWS, Azure, or OCI, including IAM and least-privilege design.
- CI/CD experience with GitHub Actions, Azure DevOps, GitLab CI, or Jenkins.
- Experience with Docker and Kubernetes.
- DataDog strongly preferred for observability tooling.
- Expert-level Python (async, API integration, testing).
- Secrets management with Vault, AWS Secrets Manager, or Azure Key Vault.
- Strong Git-based workflows and code review discipline.
Nice to have
- Experience building agents that perform write actions against production infrastructure.
- Implementation experience with Model Context Protocol (MCP) server/client components.
- Exposure to FinOps tooling such as Cloudability or Apptio.
- Oracle Cloud Infrastructure (OCI) experience.
- Healthcare or other regulated-industry environment experience.
Location: Weehawken, NJ (onsite)
Compensation: USD 84 per hour
Experience: 2+ years
Job type: contract