AI Automation Engineer / Architect of Agentic AI (Automator)

About Dizzaract

Dizzaract is a UAE-based game development studio founded in 2022, headquartered at Yas Creative Hub, Abu Dhabi. We develop cutting-edge AI-powered games and systems, including our innovation R&D laboratory FAR labs, the upcoming hero shooter Farcana, and the AI gaming identity platform GAMED. Our research and development team boasts over 100 peer-reviewed papers and more than 20 patents in AI-driven gameplay, digital ownership, and competitive design. With a diverse team of more than 80 professionals from over 20 countries, we are committed to innovation, excellence, and building a culture that drives performance and results.

Who we're looking for:We need an Automator. Not just someone who uses neural networks or a ChatGPT subscriber. We need someone who deeply understands AI at an engineering level. You'll be the person who turns scattered LLMs into a structured, hierarchical army of agents capable of running a business.

What you'll do:

• Prompt Engineering & Architecture:Develop complex, multi-layered prompts with dynamic context, few-shot learning, and self-correction mechanisms. Your code won't just be text—it will be instructions for AI employees.
• Building "Corporate Claude":Deploy and configure a corporate AI hub (based on Claude, GPT, open-source models, or local LLMs) that has access to internal documentation, knowledge bases, and company APIs.
• Agent Hierarchy Design:Design a multi-agent system. You'll create a "Director" (orchestrator) that assigns tasks to "Department Heads" (specialized agents), who then manage "Executors." You'll need to establish clear separation of roles, permissions, and responsibilities across departments (from marketing and development to legal and HR).
• Integration & Automation:Embed AI into CRM systems, databases, corporate messengers, and CI/CD pipelines. Ensure agents don't just chat—they take action: send reports, approve documents, deploy code, analyze logs.

Your skill set (what we expect):

• Deep LLM Engineering: You don't use AI blindly. You understand tokenization, temperature, logits, system prompts, and RAG (Retrieval-Augmented Generation) from the inside out. You can craft prompts that are resistant to injection attacks and hallucinations.
• Agentic Frameworks: Experience with agent-building frameworks (LangChain, LangGraph, AutoGen, CrewAI, or building custom orchestrators from scratch).
• Backend & API: Strong proficiency in Python or TypeScript. Ability to wrap agents in APIs, configure webhooks, and handle asynchronous operations.
• MCP (Model Context Protocol) or Custom Tools: Expertise in equipping AI with tools and functions. Building tools that interact with the corporate environment.
• Security: Understanding how to enforce access controls so that the sales department's agent cannot delete the development team's database.

Conditions:

• Fully remote.
• Results-oriented.
• Employment Type: Full-time.
• Salary: Competitive + KPI-based bonuses.

How to apply: please, as a portfolio share ->

• A diagram or description of a multi-agent system architecture you've built (or would like to build).
• An example of a complex prompt with function calling (tools) that you've developed.
• Your thoughts on how you would enforce access controls for agents across different departments within a corporate environment.

Job Type: Full-time

Application Question(s):

• Do you understand tokenization, logits, temperature, and their effect on LLM output behavior?
• Have you ever implemented a Retrieval-Augmented Generation (RAG) pipeline from scratch (not using a managed service)?
• On a scale of 1–5, how confident are you in building prompts resistant to injection attacks? (1 = not confident, 5 = expert)
• Rate your understanding of API authentication methods (OAuth, API keys, JWT) (1 = vague, 5 = expert)
• Have you built a self-correcting prompt with an automated feedback loop?
• Number of years of professional experience in LLM engineering (0, 1, 2, 3, 4, 5+) → (0–5)
• Is your portfolio submission complete with all three requested items (diagram, complex prompt, access control thoughts)?
• Have you built or deployed a multi-agent system (not just single-agent chatbots)?
• Have you implemented function calling (tools) where an agent executes external APIs autonomously?
• Number of production LLM automation projects you have shipped? (0, 1–2, 3–5, 6-8, 10+) → (0,1,2,3,4)
• Have you integrated LLM agents with CRM, database, or CI/CD pipelines?
• Can you build an API wrapper around an agent that handles async webhooks?
• Rate your Python proficiency (1 = basic, 5 = architect-level)
• Rate your TypeScript proficiency (1 = basic, 5 = architect-level)

Work Location: Remote