Applied AI Engineer
About SAITC
SAITC is an AI software development company in Saudi Arabia building production AI solutions including RAG systems, AI agents, enterprise chatbots, document intelligence, AI automation, and custom AI applications.
We are looking for a Senior Applied AI Engineer who can take AI capabilities from experimentation to reliable production systems.
This is an engineering role rather than a research-only role.
Role Overview
The Applied AI Engineer will own SAITC's AI engineering layer, including LLM integrations, RAG pipelines, agents, model routing, embeddings, document processing, AI evaluation, and AI-related backend services.
The ideal candidate understands not only how to use AI models, but also how to integrate, evaluate, optimize, deploy, monitor, and maintain them in production.
Key Responsibilities
LLM & Generative AI Engineering
- Design and implement production applications using large language models.
- Integrate models and APIs from providers such as OpenAI, Google Gemini, OCI Generative AI, and other relevant providers.
- Work with open-weight/open-source models such as Llama, Qwen, Mistral, and similar models where appropriate.
- Design prompts, structured outputs, tool-calling workflows, and model interaction patterns.
- Build model abstraction and routing layers that allow SAITC applications to use multiple model providers.
- Optimize model selection based on quality, latency, cost, privacy, and customer requirements.
RAG & Knowledge Systems
- Design and implement production-grade RAG pipelines.
- Build document ingestion, parsing, chunking, embedding, retrieval, reranking, and response-generation workflows.
- Work with vector databases and PostgreSQL/pgvector where appropriate.
- Build knowledge-base systems for websites, PDFs, Word documents, product catalogs, policies, and other customer data.
- Optimize retrieval quality and reduce hallucinations.
- Design multilingual retrieval systems, particularly for Arabic and English content.
AI Agents & Automation
- Design and implement AI agents and agentic workflows.
- Integrate AI agents with APIs, databases, ERP systems, CRM systems, Odoo, and business tools.
- Implement tool calling, workflow orchestration, memory, guardrails, and human escalation where appropriate.
- Determine when an agent is appropriate versus a deterministic software workflow.
AI Backend Services
- Build production APIs and services supporting AI applications.
- Collaborate closely with the Lead Software Engineer to integrate AI services into SAITC's main application architecture.
- Build background processing and document-processing pipelines.
- Implement streaming responses where appropriate for chatbot and AI applications.
- Maintain clean separation between AI logic and core business/application services.
AI Evaluation & Quality
- Establish evaluation datasets for client-specific AI systems.
- Test retrieval accuracy, response accuracy, hallucination rates, language quality, tool usage, and safety.
- Create automated AI evaluation pipelines.
- Run regression evaluations whenever models, prompts, retrieval strategies, embeddings, or system configurations change.
- Investigate and improve poor AI responses using measurable evaluation methods.
- Test Arabic and English use cases and mixed-language queries.
AI Cost & Performance Optimization
- Optimize token usage, prompt construction, caching, context size, retrieval depth, and model selection.
- Design routing strategies that use smaller models for simpler tasks and stronger models for complex tasks.
- Evaluate when self-hosted models are more economical than external APIs.
- Monitor model latency, throughput, token consumption, and inference cost.
Responsible AI & Security
- Implement safeguards against prompt injection, data leakage, unauthorized tool use, and inappropriate model behavior.
- Ensure customer information is only passed to models and services permitted by the customer's data requirements.
- Work with the Lead Engineer and DevOps Engineer on secure handling of AI credentials, data, logs, and model requests.
Required Skills
- 2-5 years of relevant professional experience in AI/ML/software engineering.
- Strong Python skills.
- Strong understanding of LLM application development.
- Production experience with RAG.
- Experience with embeddings and vector search.
- Experience using LLM APIs.
- Experience building AI applications that are deployed to production.
- Strong understanding of APIs, backend services, and software engineering fundamentals.
- Experience with prompt engineering and structured outputs.
- Experience evaluating and improving LLM systems.
Preferred Skills
- LangChain, LlamaIndex, or similar frameworks.
- Hugging Face Transformers.
- PyTorch.
- vLLM or other model-serving technologies.
- Llama, Qwen, Mistral, or other open-weight models.
- Agent frameworks.
- OCR/document intelligence.
- PostgreSQL/pgvector.
- Redis/queues.
- Docker.
- FastAPI.
- Cloud AI services, particularly OCI, GCP, or AWS.
- Experience with Arabic NLP or multilingual AI.
- Knowledge of LLM fine-tuning, quantization, distillation, or inference optimization.
What Success Looks Like
The engineer should be able to take a requirement such as:
"Build a bilingual AI customer-service assistant that understands our documents, answers customer questions, connects to our CRM, and escalates complex cases."
and independently turn it into a tested, measurable, deployable AI system in collaboration with the rest of the engineering team.
Pay: ₹60,000.00 - ₹70,000.00 per month
Education:
Experience:
- AI: 2 years (Preferred)
- Software development: 3 years (Preferred)
Work Location: In person
