Free shipping on orders over 1,300 EGP العربية
Free advice from licensed male & female pharmacists on WhatsApp — ask now
AI & Automation Engineer

Recent Viewed Products

Tech & AI

AI & Automation Engineer

Agents that know when to ask a pharmacist

Always hiringMid-levelFull-timeOn-siteNew Cairo, Egypt
Share this role: LinkedIn WhatsApp

Every day at Beseha, the same customer questions come in, product descriptions wait to be written, and orders get moved by hand between the store and Odoo. You will turn that work into tools and agents built on large language models like Claude, tools that run in production, are measured in numbers, and have guardrails that keep them from getting health information wrong.

Experience: 2-5 years of software development, including hands-on experience building applications with large language models (or equivalent shipped real-world projects)

Why this role mattersWhen a store sells supplements and personal care, a wrong answer is more than a bad experience; it can affect someone's health. We need an engineer who uses AI to speed up our work and widen access to trustworthy health information, while respecting pharmacist review and the limits of what a machine should say.

What you will do

  • Build LLM-powered agents and tools that automate content writing, review summarisation and on-site search, so the team can focus on work that needs human judgement.
  • Design suggested replies and customer service assistants that know their limits and route sensitive medical questions to a pharmacist, so customers get answers that are both fast and safe.
  • Write and refine prompts and retrieval logic over product data and health guides (RAG), so outputs are grounded in Beseha's verified sources rather than the model's guesses.
  • Create test sets and recurring evaluation criteria (accuracy, safety, tone, cost) and rerun them after every change, so every improvement is proven with data.
  • Integrate tools with WooCommerce and Odoo through APIs and webhooks, so automation reaches real execution in orders and inventory.
  • Put clear quality and safety controls in place: human review where it matters, output logging, and protection of customer data.
  • Monitor cost and latency and choose the right model for each task, so solutions stay sustainable as usage grows.
  • Document what you build and train the teams who use it, so tools are genuinely adopted instead of staying side experiments.

What we need

  • 2-5 years of software development, including hands-on experience building applications with large language models (or equivalent shipped real-world projects).
  • Practical skill in Python or JavaScript/TypeScript, including building integrations over REST APIs.
  • At least one shipped or actively used project built on an LLM (Claude, GPT or similar) that you can demo or walk us through.
  • Strong prompt design and tool use / function calling, with a working grasp of RAG fundamentals.
  • Experience evaluating model outputs systematically: test sets, metrics and version comparisons.
  • Ability to read and write Arabic and technical English, since much of the output is in Arabic.

Nice to have

  • Experience with WordPress/WooCommerce, Odoo, or any e-commerce platform or ERP.
  • Familiarity with automation tools such as n8n or Make, or with agent frameworks.
  • Using BigQuery or SQL to analyse how tools perform.
  • Interest in health and nutrition, or prior work on a product that handles sensitive content.

Your first 90 days

  • By day 30You understand how work flows through content, customer service and operations, have agreed with the team on the two highest-impact use cases, and have built a first test set for each.
  • By day 60You have shipped the first tool to real users (for example, product description drafts or suggested customer service replies) with human review, clear safety guardrails and a quality dashboard.
  • By day 90That first tool is connected to the store or Odoo and saves measurable time, you run a documented recurring evaluation, and you have a written plan for the next use case.

Where this can lead

  • Senior AI Engineer, owning agent architecture and evaluation across the company
  • Lead of an AI & Automation team
  • A specialist path in AI product, or in model safety and evaluation

Know someone who fits this role?

Make a referral link with your name and send it. Anyone who applies from it is recorded as coming from you.

Anyone who applies from your link is recorded as referred by you. We keep only the name, for follow-up.

About Beseha

Beseha is building a healthier Egypt: pharmacist-reviewed products, reliable health information, and a service that respects people. We look for different people to grow with us and grow that impact. Read our mission

More roles in this team

0
Ask a pharmacist — free Available now · Male & female pharmacists · reply in 5 min