Applied AI Engineer - Agentic Systems

DeepLight AI is a specialist AI and data consultancy dedicated to transforming the regional corporate landscape through bespoke, high-impact intelligent systems. Based in the UAE, we partner with major enterprise and financial services clients to bridge the gap between cutting-edge AI research and operational business reality.

We don't build isolated prototypes or generic "off-the-shelf" wrappers. We engineer secure, scalable, and resilient AI foundations designed to integrate seamlessly into complex corporate environments.

The Opportunity We are seeking a hands-on Applied AI Engineer – Agentic Systems to design, build, and deploy LLM-powered autonomous agents and multi-step agentic workflows for real-world enterprise use cases.

In this role, you will bridge the gap between advanced AI capabilities and production software engineering. You will combine tool calling, state management, context engineering, and external API integrations to transform complex business processes into reliable, auditable, and maintainable AI applications.

What You’ll Do Architect Autonomous Workflows: Design and implement multi-step agent loops, state machines, and tool-calling mechanisms that execute complex enterprise tasks autonomously and reliably. Master Context Engineering & RAG: Build advanced context management pipelines and Retrieval-Augmented Generation (RAG) modules to ground agent reasoning in dynamic enterprise knowledge sources. Integrate & Connect Systems: Develop secure API integrations connecting LLMs and agents to enterprise databases, legacy line-of-business applications, and modern protocol standards (such as MCP). Engineering for Production & Fail-Safes: Write clean, production-ready Python code backed by automated testing, structured output validation (e.g., Pydantic), fallback logic, and human-in-the-loop (HITL) oversight controls. Evaluate, Trace & Optimize: Instrument end-to-end tracing and evaluation suites (e.g., LangSmith, Phoenix) to benchmark performance, eliminate hallucination loops, diagnose latency spikes, and optimize token costs. Translate & Advocate: Collaborate directly with client technical teams and business sponsors to translate ambiguous processes into clear technical designs, acceptance criteria, and system architectures.

As an AI consultancy, our greatest asset is the expertise of our people.

While technical mastery is the foundation of what we do, the ability to bridge the gap between complex data science and actionable business value is what defines your success with Deeplight.

We're looking for individuals who are not only world-class in their fields of specialism, but also compelling communicators and persuasive advocates for their own skills.

You will be the face of our firm, tasked with building trust, articulating the "why" behind your technical decisions, and effectively "selling" your vision to high-level stakeholders.

If you thrive on the challenge of presenting cutting-edge solutions as much as you do on building them, you will fit right in.

Requirements

What We’re Looking For Demonstrable experience personally implementing AI agents, multi-agent frameworks, tool calling, and state management systems. Advanced proficiency in Python with a focus on code maintainability, testing (pytest), code reviews, and Git version control. Practical experience integrating commercial and open-weight LLM APIs, designing prompts, validating structured outputs, and gracefully handling model/tool failures. Experience evaluating LLM applications using representative task sets, failure mode analysis, regression testing, and application tracing tools. Ability to articulate and balance decisions across reasoning quality, system reliability, inference latency, and operational cost. What Gives You an Edge Experience deploying and supporting LLM agents used by active business teams or external customers. Familiarity with modern orchestration frameworks (e.g., LangGraph, AutoGen, CrewAI, LlamaIndex Workflows) or custom state-machine implementations. Hands-on experience with the Model Context Protocol (MCP), Docker containerization, and cloud deployment (AWS/Azure). Prior experience delivering technical solutions directly within regulated enterprise environments (e.g., Banking, Financial Services, Insurance). Demonstrating Your Expertise We value practical, battle-tested engineering. During our process, candidates should be prepared to walk through a system they personally built-explaining implementation choices, evaluation approaches, system limitations, and how they navigated the journey from prototype to production.

Benefits

The benefits you'll enjoy as part of this role include:

Competitive salary Comprehensive 'Gold Level' personal health insurance Individual Visa Sponsorship Professional development and certification support Subscription reimbursement relating to your role Opportunity to work on cutting-edge AI projects Monthly Employee Incentive program Career advancement opportunities in a rapidly growing AI company This position offers a unique opportunity to shape the future of AI implementation while working with a talented team of professionals at the forefront of technological innovation. The successful candidate will play a crucial role in driving our company's success in delivering transformative AI solutions to our clients.

At DeepLight AI, we recognise that diversity drives innovation. We are committed to fostering an inclusive environment where individuals with different thinking styles can thrive and contribute their unique strengths to our specialised AI and data solutions.

Our goal is to ensure our application and interview process is accessible, predictable, and fair for all candidates.

If you require any specific adjustments to the application process, or if you require any reasonable adjustments should you be successful in being processed to the interview stage, please do let us know. This information will be kept strictly confidential and will not impact hiring decisions.