Senior ML Platform Engineer, Inference & MLOps

Mollkom; Riyadh; Hybrid; SAR 150,000 - 210,000; Full time

Job summary

Mollkom is seeking an AI infrastructure engineer to design serving, routing, and concurrency controls. You will optimize caching, batching, and token use while measuring output quality and cost. The role emphasizes observability, model/versioning, and progressive rollout strategies.

Join Mollkom in Riyadh for a hybrid work setup, focusing on building scalable, low-latency AI services with robust failure handling and CI/CD automation.

Qualifications

• Experience in ML infrastructure, MLOps or operating AI services.
• Strong Python, Go or TypeScript and understanding of containers and cloud services.
• Experience with observability, CI/CD, secrets and failure handling.
• Ability to locate bottlenecks with measurements and explain performance/cost trade-offs.

Responsibilities

• Design serving, provider routing, queues and concurrency controls.
• Optimize caching, batching and token use while measuring output quality.
• Build latency, error and cost observability with model/configuration versioning.
• Develop progressive rollout, recovery and replay for long-running work.

Skills

ML Infrastructure

Python

Go

TypeScript

Cloud services

Observability

CI/CD

Containers

Tools

Docker

Kubernetes

CI/CD pipelines

Job description

AI workflows need precise operational choices, from routing and scheduling to measurement and versioning. Develop infrastructure that lets teams experiment and ship while understanding quality, latency and cost.

Responsibilities

• Design serving, provider routing, queues and concurrency controls.
• Optimize caching, batching and token use while measuring output quality.
• Build latency, error and cost observability with model/configuration versioning.
• Develop progressive rollout, recovery and replay for long-running work.

Qualifications

• Experience in ML infrastructure, MLOps or operating AI services.
• Strong Python, Go or TypeScript and understanding of containers and cloud services.
• Experience with observability, CI/CD, secrets and failure handling.
• Ability to locate bottlenecks with measurements and explain performance/cost trade-offs.

Hybrid in Riyadh, Saudi Arabia.