Mohamed Maktal

About Mohamed Maktal

I’m a Solutions Engineer at Vultr helping customers ship AI/GPU and cloud infrastructure, from first discovery call to production handoff. My sweet spot is translating business goals into architectures that are fast, cost-efficient, and reliable.What I do:• Sales engineering: discovery → architecture → POC → close. I quantify TCO/perf trade-offs and unblock security, networking, and compliance.• AI/GPU: design GPU stacks for training/inference, K8s-native MLOps (containers, registries, schedulers), perf validation, and right-sizing.• Cloud/CPU: bare metal & cloud compute, VPC/BGP/VPN, storage patterns, IaC (Terraform), CI/CD, monitoring & SLOs.Backstory: With over a decade in cloud, I started in support in the voice/UC space (Shifa → 8x8 → Twilio) learning how to run production, customer-facing systems at scale. That customer empathy now powers my work in infra at Vultr with the same obsession, uptime and outcomes.Topics of interest: GPU infra, K8s, cost/perf tuning, and practical SE playbooks.Let’s connect if you’re designing AI/GPU clusters, migrating workloads, or need an honest TCO/performance view.

Experience

1. Solutions Engineer

   Vultr

   Aug 2024 — Present · Seattle, WA, US

   Lead discovery, architecture, POC/benchmarking to support our account execs. Quantifying TCO vs. performance and aligning stakeholders (infra, security, finance) to accelerate decisions while reducing the sales cycle time.Architected scalable infrastructure for AI/ML training and inference workloads, utilizing the latest and greatest NVIDIA & AMD GPUs while advising clients on cost-efficient scaling with an average of 30-40% in savings regularly.Design and implement multi-region, highly available (HA) solutions using Kubernetes (VKE), high-frequency CPUs, and Cloud Block Storage, ensuring 99.99% uptime for our customer\'s critical applications.

Education

• University of Minnesota

  BS, Computer Science