AI Solutions Architect

Clients ask us for AI. Some need it, some need a search box and a tidier database, and most need a smaller version of what they asked for. You're the person who works that out in the room, with their data on screen.

This is client-facing architecture: discovery workshops, feasibility against real data, a design our engineers can build, and the business case that decides whether it goes ahead at all.

You stay involved through delivery, which means you find out quickly when a design doesn't survive contact with the client's actual data. That feedback loop is the point.

The studio runs on agents itself, so when you tell a client what's possible, you're describing how we already work, not something you read about.

• Run discovery properly. Business and technical people in the same workshop, ending with a short list worth building.
• Test feasibility early. Against their real data, not the version described confidently in the meeting.
• Design the whole thing. Retrieval, models, integration, human review, and what happens on the day the model is wrong.
• Put numbers on it. Cost per request, expected quality, and what has to be true for the business case to hold.
• Write the proposal. And defend it, including the version where the recommendation is to wait.

• Five Years in Software, Recent AI Delivery: hands-on, in production, not advisory slides about someone else's build.
• Reference Architectures You Have Drawn: retrieval, agents, batch scoring, and the unglamorous integration layer underneath all of it.
• Model Landscape Awareness: hosted and open weight, with the cost, latency and data residency argument for each.
• Data Readiness Assessment: you can tell a client their documents aren't ready, and show them what ready would look like.
• Cloud Platform Fluency: Bedrock, Azure OpenAI, Vertex or Cloudflare, and what each one locks you into.
• Compliance Conversations: personal data, residency, retention and audit, discussed with their risk officer without flinching.
• Build Versus Buy Calls: you say so when a product off the shelf beats anything we could bill for.
• Systems That Shipped: you can describe how one failed and what you changed because of it.
• Bilingual, Technically: credible with engineers, useful to a CFO, in the same afternoon.
• Estimator With A Spine: you size effort and cost, and hold the line when the client wants the number smaller.
• Strong Writer: proposals, architecture notes and the one-page summary an executive will read.
• Comfortable Saying No: with evidence, in a way that keeps the relationship.
• Fluent English, Written and Spoken: most of our work happens in a written thread or on a call, and both have to be clear. Arabic is a genuine advantage across the region.
• Practices What You Propose: you use these tools daily, so your feasibility estimates come from experience instead of a vendor deck.

• Commercially minded: you think about what this costs the client, not only what it does.
• Skeptical of hype: including your own field's.
• Persuasive: you bring people with you rather than winning on volume.

• Regulated industries and the constraints they put on AI.
• Presenting to boards or investment committees.
• A background in data science or data engineering.

• Remote across MENA. Work from wherever you are in the region, with hours built around your team's overlap rather than someone's attendance spreadsheet.
• Work that ships. Products go live in weeks, not quarters. When someone asks what you have been building, you can open it on your phone and show them.
• Small teams, no ticket queue. Four to eight people carry a product from the first client call to handover. Nobody hands you a spec and disappears.
• Kit and tools you choose. A budget for your setup, and for the editor, licenses and services that make you fast. No three-week procurement saga.
• A learning budget with time attached. Money for the course or the conference, plus room in the week to use it. The second half is the part most places skip.
• Client contact from day one. You sit in the room where the decisions get made. Plenty of engineers wait a decade for that seat.
• Time off people take, and no timesheets. We measure engagements, not humans in fifteen-minute increments. Book the holiday, go, come back.