Ai engineer Software Developer

AI Engineer – Job ProfileJob Title

Senior AI Engineer – LLM, RAG, Computer Vision & AI Agents

Department

Artificial Intelligence & Emerging Technologies

Job Location

UAE – Abu Dhabi / Dubai

Job Summary

We are looking for an experienced AI Engineer to design, develop, deploy, and maintain advanced artificial intelligence solutions for enterprise and government projects.

The candidate will work across Generative AI, Large Language Models, RAG, AI Agents, Computer Vision, OCR, Speech AI, multimodal AI, machine learning, and private/on-premise AI platforms.

The position requires a strong practical engineering background and the ability to take an AI solution from concept through development, integration, testing, deployment, optimization, and production support.

1. Large Language Models & Generative AI

The candidate will be responsible for:

• Developing applications using Large Language Models and Small Language Models.
• Building enterprise AI assistants and copilots.
• Developing AI-powered document analysis applications.
• Creating automated RFP and tender analysis solutions.
• Developing technical proposal generation systems.
• Implementing AI knowledge assistants.
• Building compliance and engineering AI assistants.
• Implementing prompt engineering and prompt optimization.
• Implementing structured outputs and function calling.
• Developing model-routing mechanisms.
• Managing context windows and conversation memory.
• Integrating commercial and open-source AI models.

Preferred model experience includes:

• Qwen
• Llama
• Gemma
• Mistral
• DeepSeek
• OpenAI-compatible models
• Vision-Language Models
• Multimodal models

2. Retrieval-Augmented Generation – RAG

The candidate should have experience building enterprise RAG systems capable of processing:

• PDF documents.
• Microsoft Word documents.
• Excel spreadsheets.
• PowerPoint presentations.
• Scanned documents.
• Images.
• Technical manuals.
• Contracts.
• Policies.
• Standards.
• Tender and RFP documents.
• Engineering documentation.

Responsibilities include:

• Document ingestion.
• Text extraction.
• OCR processing.
• Document parsing.
• Intelligent chunking.
• Metadata extraction.
• Embedding generation.
• Semantic search.
• Keyword search.
• Hybrid search.
• Query rewriting.
• Reranking.
• Context construction.
• Citation generation.
• Retrieval optimization.
• RAG evaluation.
• Hallucination reduction.

Preferred vector database experience includes:

• Qdrant.
• Milvus.
• Weaviate.
• FAISS.
• PostgreSQL / pgvector.
• Elasticsearch.
• OpenSearch.

3. AI Agents & Multi-Agent Systems

The candidate will design AI agents capable of:

• Calling APIs.
• Accessing databases.
• Reading enterprise documents.
• Searching internal knowledge.
• Executing workflows.
• Performing calculations.
• Generating reports.
• Using business applications.
• Calling external tools.
• Validating generated results.

The candidate should understand:

• Tool calling.
• Function calling.
• Planning.
• Reasoning workflows.
• Agent memory.
• Agent orchestration.
• Human-in-the-loop workflows.
• Multi-agent systems.
• Supervisor agents.
• Worker agents.
• Research agents.
• Validation agents.
• Reporting agents.
• Specialized engineering agents.

Preferred frameworks include:

• LangChain.
• LangGraph.
• LlamaIndex.
• Semantic Kernel.
• CrewAI.
• AutoGen.
• Custom agent frameworks.

4. Computer Vision

The candidate should have practical experience developing Computer Vision applications covering:

• Object detection.
• Object classification.
• Image segmentation.
• Object tracking.
• Person detection.
• Vehicle detection.
• PPE detection.
• Intrusion detection.
• Restricted-area detection.
• Loitering detection.
• Crowd analysis.
• Fire detection.
• Smoke detection.
• Defect detection.
• Asset inspection.
• Road inspection.
• CCTV analytics.
• Image analysis.
• Video analytics.

Preferred technologies include:

• YOLO.
• OpenCV.
• PyTorch.
• TensorFlow.
• Detectron.
• Segment Anything.
• Vision Transformers.
• Vision-Language Models.

The candidate should understand:

• RTSP video streams.
• Real-time video processing.
• Edge inference.
• GPU-based inference.
• Image preprocessing.
• Video frame processing.
• Object tracking.
• Event detection.
• Alert generation.

5. Multimodal AI

The candidate should be capable of building AI solutions combining:

• Text.
• Images.
• Video.
• Audio.
• Documents.
• Database information.
• IoT data.
• Sensor information.

Typical multimodal workflows may include:

• Image → Vision Model → LLM → Business Rules → Recommendation.
• CCTV → Computer Vision → Event Detection → LLM → Alert.
• Document → OCR → RAG → LLM → Compliance Engine.
• Voice → Speech-to-Text → LLM → Business Application.
• Mobile Camera → Vision AI → Defect Detection → Inspection Report.

6. Speech AI

The candidate should have experience with:

• Speech-to-Text.
• Text-to-Speech.
• Voice assistants.
• Audio processing.
• Arabic speech recognition.
• English speech recognition.
• Multilingual voice applications.
• Voice-controlled AI systems.

Preferred technology experience includes:

• Whisper.
• wav2vec2.
• Vosk.
• Coqui.
• Piper.
• Commercial speech APIs.

Arabic AI and speech experience is highly desirable.

7. OCR & Document Intelligence

The candidate will develop intelligent document-processing solutions covering:

• OCR.
• Arabic OCR.
• English OCR.
• Table extraction.
• Form extraction.
• Document classification.
• Layout analysis.
• Data extraction.
• Key-value extraction.
• Document understanding.
• Validation workflows

Education:

• Bachelor's (Preferred)

Experience:

• AI, machine learning, or data science: 3 years (Preferred)
• LLMs, RAG, AI Agents, or Computer Vision: 3 years (Preferred)
• AI solution into a production environment: 1 year (Preferred)
• LLM Development : 3 years (Preferred)

License/Certification:

• Driving license (Preferred)

Location:

• Abu Dhabi (Preferred)

Work Location: In person