Senior Agentic AI & LLM Engineer

imkan.ai; Riyadh; Vor Ort; SAR 420.000 - 660.000; Vollzeit

Zusammenfassung

imkan.ai is seeking a senior AI/ML lead to design and deploy production-grade agentic AI and LLM systems for enterprise clients across Saudi Arabia. You will own multi-agent architectures, RAG pipelines, and end-to-end integrations from architecture through live deployment, collaborating with imkan.ai’s engineering leaders.

Role requires at least 6+ years delivering production AI, strong LLM experience, and knowledge of governance, observability, and cost optimization.

Qualifikationen

• 6+ years of hands-on experience delivering production AI or data systems.
• Hands-on with LLMs: prompt engineering, fine-tuning, RAG, model compression.
• At least one agentic AI system taken from design through to production.
• Familiarity with AI model risk frameworks and audit trail requirements.
• Experience delivering AI solutions in Saudi Arabia or the wider GCC region is a strong advantage.

Aufgaben

• Lead the design, development, and deployment of production-grade agentic AI and LLM systems.
• Own multi-agent architectures, RAG pipelines, and LLM integrations end-to-end.
• Build and maintain data pipelines, vector store architectures, and embedding workflows.
• Manage cloud infrastructure (Azure/AWS) with cost optimization and governance.
• Define observability standards including logging, tracing, and drift detection in production.

Kenntnisse

PyTorch

LangGraph

Transformers

LangChain

AutoGen

CrewAI

PySpark

Airflow

Kafka

Flink

Azure

AWS

GCP

Tools

dbt

PySpark

Airflow

Kafka

Flink

Jobbeschreibung

imkan.ai is an international Data and AI consultancy headquartered in Saudi Arabia, specialising in engineering-led solutions across data infrastructure and agentic AI systems. We partner with global enterprises to move data and AI from concept to production, building things that actually run.

ABOUT THE ROLE

You will lead the design, development, and deployment of production-grade agentic AI and LLM systems for enterprise clients across Saudi Arabia. This is not a research role. You will own multi-agent architectures, RAG pipelines, and LLM integrations end-to-end, from architecture through to live deployment. You will work embedded with client teams and in close coordination with imkan.ai’s engineering leads.

KEY RESPONSIBILITIES

• Build and maintain production-grade data pipelines to support LLM ingestion, embedding generation, and retrieval workflows
• Design vector store architectures and manage embedding pipelines across structured and unstructured enterprise data sources
• Manage cloud AI infrastructure on Azure or AWS, including model hosting, API gateway management, and cost optimisation
• Enforce data quality, lineage, and governance standards relevant to enterprise AI deployments in regulated environments
• Define observability standards for AI systems including logging, tracing, drift detection, and cost monitoring in production

Agentic AI & Generative AI

• Build and maintain end-to-end AI orchestration pipelines covering data ingestion, model inference, and result delivery
• Design and operate RAG pipelines with vector databases for enterprise knowledge retrieval
• Engineer, version, and continuously refine prompts for production LLM applications
• Integrate LLMs with external tools and APIs to build autonomous, multi-step agentic workflows
• Implement model evaluation frameworks, guardrails, and safety checks for enterprise LLM deployments across client environments
• Lead incident triage and root-cause analysis for AI-driven services in production
• Collaborate with data scientists, product owners, and platform engineers across time zones

REQUIRED SKILLS & EXPERIENCE

Programming & Frameworks

• AI/ML: PyTorch, Scikit-learn, Hugging Face Transformers, LangChain, LangGraph
• Agentic frameworks: LangGraph, AutoGen, CrewAI, or similar multi-agent orchestration tools
• Data tooling: dbt, PySpark, Airflow, Kafka or Flink
• Cloud: Azure (preferred), AWS, or GCP

Domain Experience

• 6+ years of hands-on experience delivering production AI or data systems, not PoC or academic projects
• Hands-on with LLMs: prompt engineering, fine-tuning, RAG, model compression
• At least one agentic AI system taken from design through to production
• Familiarity with AI model risk frameworks and audit trail requirements
• Experience delivering AI solutions in Saudi Arabia or the wider GCC region is a strong advantage

Ways of Working

• Fluent English, written and spoken. This is the working language across all teams
• Currently based in Saudi Arabia with a valid, transferable Iqama. This is a non-negotiable requirement
• Agile/Scrum proficiency; able to self-manage against sprint commitments

NICE TO HAVE

• Familiarity with Saudi data and AI governance frameworks, NCA requirements, or PDPL compliance considerations
• MLflow or equivalent for experiment tracking and model registry