AI Engineer

Company Description: Pioneers Information Technology Co. Ltd., part of Al-Nafie Holding Group, is a technology solutions provider founded in 2011 and headquartered in Jeddah, KSA, with additional branches in Riyadh, Cairo, and Dubai. As a Microsoft Gold Certified Partner and Tier 1 Cloud Solution Provider, the company delivers professional services in development, implementation, training, and support of integrated business management systems. Pioneers IT specializes in Microsoft Dynamics ERP, CRM, Power BI, and industry-focused add-ons such as Van Sales, E-Invoicing, Localized Payroll, Property Management, and Student Registration. Backed by a team with extensive experience in technology and ERP, the company focuses on enabling successful digital transformation for clients through customized, scalable solutions. Ongoing investment in certifications and skills ensures that specialists stay current with the latest advancements in information technologies.

Role Description

This person owns hands-on, end-to-end development of AI agents across our agents— building the orchestration, retrieval, guardrails, and evaluation layers that turn raw Dynamics data into reliable, plain-language business intelligence. They are a builder, not a researcher: the work is LLM orchestration over existing enterprise data, not custom model training.

Key Responsibilities

•Design, build, and ship custom engine agents using LangGraph / LangChain — from prototype through production — bringing our own orchestration rather than relying on Copilot's built-in orchestrator.

• Integrate LLMs through Microsoft Foundry (Claude and GPT as swappable engines) rather than calling providers directly; manage prompts, tool definitions, and function calling.

• Implement RAG pipelines — vector stores, retrieval, grounding agent outputs in F&O and CRM data.

• Build the guardrailed, policy-gated architecture that defines these agents: a policy gate before every LLM call, plain-language outputs with citations, and a validation layer that checks answers before they reach the user.

• Track and optimize token/credit consumption to protect the SaaS unit economics — instrument per-customer (per-tenant) usage metering, monitor cost-per-query and cost-per-customer against revenue, and surface consumption in a way that ties directly to margin.

• Engineer cost efficiency into the agents — apply techniques such as prompt and context optimization, caching, model right-sizing (routing cheaper models where quality allows), retrieval tuning, and usage limits/throttling — so that no customer becomes unprofitable and gross margin is preserved as the product scales.

• Own the full lifecycle — design, prototype, evaluate, deploy, maintain — including writing test cases and defining what correct looks like for each agent.

• Deploy and operate on Azure (Foundry, Container Apps, Key Vault) — comfortably, or with the ability to learn fast alongside the Technical Consultant

Qualifications

• 3+ years software engineering experience, with strong Python (non-negotiable).

• 1–2+ years hands-on with LLMs and modern AI tooling — real work, not courses or tutorials.

• Direct experience with LangGraph / LangChain and building agentic workflows.

• RAG / vector-store experience.

• Has shipped at least one real AI product or feature and can walk through the engineering decisions behind it.

• Awareness of LLM cost drivers (tokens, context size, model selection) and how engineering choices affect them.

Nice to Have (Preferred)

• Dynamics 365 / F&O & CRM knowledge.

• Front-end / UX experience.

• Azure deployment experience (Foundry, Container Apps, Key Vault).

• Evaluation / LLM-testing discipline.

• Multi-tenant SaaS exposure and cost/FinOps awareness (per-tenant metering, unit economics).