tamm تعلن عن وظيفة مهندس أبحاث ذكاء اصطناعي في السعودية

About Tamm Tamm is the all-in-one platform for selling online in Saudi Arabia. Sellers get everything they need in one place: a storefront, payments, ZATCA-compliant e-invoicing, and delivery. And because we run our own print-on-demand factory in Dammam, anyone can launch a brand from scratch with no inventory and no capital. You upload a design, we make the product only once it sells, then we print, pack, and ship it. No warehouse, no upfront cost, no risk. Turning a design into a real business, on demand, is what we are built for.

What you'll build You'll own the model layer that makes Tamm's AI production-ready. You'll fine-tune, evaluate, and deploy models for real use cases across the product: the seller assistant, generative design and copy, and the agentic store builder. Your job is to take models, frontier and open, and make them fast, reliable, and good enough to ship.

• Fine-tune, evaluate, and deploy models for real product use cases
• Build and own the eval harness that keeps quality from regressing as we ship
• Optimize the model layer for quality, latency, and cost
• Work close to the product, so your research always ends in something shipped

What we're looking for

• Applied ML / LLM engineer with real fine-tuning and deployment experience
• Strong in Python and the ML stack (PyTorch, Hugging Face, and similar)
• Evaluation-first: you measure before you ship
• Production-focused, not paper-first
• Comfortable across the model lifecycle: data, training, serving, monitoring

Nice to have RAG and retrieval systems; inference optimization; model hosting on AWS / Kubernetes; enough TypeScript to integrate with the product.

How we work Small, senior, and fast. You own outcomes, not tickets. AI does the heavy lifting in how we build, so there's little busywork and few meetings. 100% remote, full-time.