AI Agent Deployment Engineer

SimpleCiti Companies; United States; On-site; USD 150,000 - 210,000; Full time

Job summary

SimpleCITI Companies seeks an AI Agent Deployment Engineer to lead design and deployment of autonomous, multi-agent systems across the organization. The role focuses on true agentic behavior, not simple automation, enabling agents to join meetings, communicate with tenants, support acquisitions, and manage complex tasks with minimal human input.

The ideal candidate will live and breathe AI, track framework updates, and push the frontier of agent technology.

Qualifications

• Hands-on expertise with LangChain and LangGraph for multi-agent orchestration.
• Strong Claude and Anthropic API experience for agentic workflows.
• Production experience with Hermes and open-weight LLMs.
• Experience building real tool integrations with external APIs and services.

Responsibilities

• Design, deploy, and orchestrate multi-agent systems for complex tasks.
• Architect agent layers where agents collaborate and delegate to others.
• Fine-tune models and ensure on-premise or custom behavior as needed.
• Build agents with robust tool use and long-context capabilities.

Skills

Agent orchestration

Python programming

Async programming

System design

Tools

LangChain

LangGraph

Claude

Anthropic API

Hermes

Claude Code

Paperclip

Obsidian

API design

Job description

SimpleCITI Companies, Hiring for AI Agent Deployment Engineer

Overview: SimpleCITI Companies is a vertically integrated real estate and specialty finance platform built to execute across the full investment lifecycle. Our operating companies specialize in real estate acquisitions, capital markets, brokerage, property management, development, and specialty lending. With expertise in industrial, commercial, mixed-use, and residential real estate, SimpleCITI operates at the intersection of capital deployment, operational execution, and long-term value creation.

Job Description

SimpleCITI is hiring an AI Agent Deployment Engineer to lead the design, deployment, and orchestration of true agentic AI systems across every function of the organization. This is not a role for building simple automations or basic chatbots. We are deploying autonomous, multi-agent systems that operate as full digital employees — joining meetings, making and receiving calls, managing tenant relationships, supporting acquisitions, handling investor communications, and executing complex long-horizon tasks with minimal human intervention.

The ideal candidate lives and breathes AI. They are reading agent news daily, tracking every framework update, and staying ahead of a field that changes by the week. They understand the difference between automation and true agentic behavior — and they know how to build, deploy, and manage the latter at scale.

Core Responsibilities

• Design, build, and deploy multi-agent systems using LangChain, LangGraph, and related frameworks — with a focus on true agentic behavior, not simple linear automation
• Architect agent orchestration layers where agents manage, delegate to, and collaborate with other agents to complete complex, multi-step tasks
• Deploy and fine-tune models using Hermes and other open-weight LLMs for specialized internal use cases requiring on-premise or custom model behavior
• Build and manage agents using Claude, Claude Code, and the Anthropic API — leveraging extended thinking, tool use, and long-context capabilities
• Implement memory architectures — short-term, long-term, and episodic — so agents maintain context across sessions and interactions
• Deploy voice and telephony agents capable of making and receiving calls, conducting conversations, and executing tasks through phone-based interactions
• Build agents with robust tool use — including web browsing, code execution, API calls, database queries, document generation, and email/calendar management
• Use Paperclip for agent task management, workflow tracking, and internal operational coordination
• Maintain agent knowledge bases, documentation, and second-brain infrastructure in Obsidian
• Monitor agent performance, identify failure modes, implement self-correction loops, and continuously improve reliability and output quality
• Stay ahead of the field — reading agent research, framework updates, and AI news daily — and proactively bring new capabilities into production

Requirements

• Deep, hands‑on expertise with LangChain and LangGraph — including agent graph design, state management, and complex multi‑agent orchestration
• Strong proficiency with Claude and the Anthropic API, including tool use, prompt engineering, and Claude Code for agentic development workflows
• Experience deploying and fine‑tuning Hermes and other open‑weight LLMs for production use cases
• Experience building agents with real tool use — not toy examples, but production‑grade integrations with external APIs, databases, and services
• Familiarity with voice agent platforms and telephony integrations for inbound and outbound call handling
• Proficiency with Paperclip for workflow and task management in agentic contexts
• Working knowledge of Obsidian for knowledge management, agent memory systems, and documentation infrastructure
• Strong software engineering fundamentals — Python, API design, async programming, and system architecture
• Experience building agents that operate over long time horizons — tasks spanning hours, days, or weeks with minimal human checkpointing

Bonus

• Background in RAG pipelines, vector databases, and semantic search for agent memory and retrieval
• Experience with agent evaluation frameworks and systematic approaches to testing and improving agent reliability
• Familiarity with real estate operations, PropTech platforms, or financial services workflows
• Prior experience deploying agents in enterprise or regulated environments
• Contributions to open‑source agent frameworks or published work in the agentic AI space