Mustapha Saeed

Engineered and developed the enterprise service bus (ESB) layer for a large telecommunications platform, enabling reliable integration and communication between multiple enterprise systems. The solution followed service-oriented architecture (SOA) principles and used JMS for asynchronous messaging and service integration. I contributed to the architecture, implementation, and reliability of the integration layer supporting business-critical telecom services.

Contributed to the architecture and development of a large-scale microservices platform built using Netflix OSS and deployed on Amazon ECS. I focused on building highly available and resilient infrastructure for distributed services, including deployment, monitoring, scalability, and operational reliability. The platform achieved 99.99% availability while supporting a large global user base.

Led multiple AWS cost optimization initiatives for a large-scale news crawling and data processing platform. Redesigned and optimized cloud infrastructure, compute workloads, and scaling strategies to improve resource utilization while maintaining platform reliability. These initiatives delivered more than $2 million in annual AWS savings while maintaining over 99.9% availability across the hosted applications.

Architected and operated a large-scale multi-cloud Kubernetes platform for GPU-intensive AI workloads, scaling to 1,000–2,000 nodes during peak demand. I was responsible for the platform's architecture, scalability, reliability, and GPU capacity across multiple cloud providers and regions. I focused heavily on maximizing GPU utilization, maintaining capacity availability, automating cluster operations and scaling, and optimizing infrastructure costs while supporting highly dynamic AI workloads.

Designed and implemented a comprehensive observability platform using the LGTM stack (Loki, Grafana, Tempo, and Mimir), providing centralized visibility across infrastructure and applications. Built dashboards, alerting, log aggregation, distributed tracing, and metrics monitoring to improve platform reliability and incident response. The entire observability stack was deployed and managed through Terraform, ensuring a consistent, scalable, and reproducible infrastructure.