Principal Software Engineering Manager - Substrate Efficiency - Microsoft Corporation

Key Responsibilities Lead and mentor a multidisciplinary engineering group responsible for the core LLM inference substrate powering Microsoft 365 Copilot. Design, implement, and optimize low‑latency, high‑throughput GPU‑accelerated services that route and serve massive AI workloads across global datacenters. Drive architecture decisions for distributed systems, ensuring fault tolerance, scalability, and observability at massive scale. Collaborate with product, research, and infrastructure teams to integrate new AI models and features while maintaining strict performance SLAs. Establish best practices for CI/CD, monitoring, and ML‑ops pipelines on Azure, fostering a culture of reliability and continuous improvement. Requirements 10+ years of software engineering experience, with at least 5 years in a leadership role building large‑scale, performance‑critical systems. Deep expertise in C++ and Python for systems and AI workloads, plus hands‑on experience with GPU programming (CUDA, ROCm) and high‑performance networking. Proven track record designing distributed, fault‑tolerant architectures on cloud platforms, preferably Azure. Strong understanding of large language model serving, inference optimization, and ML‑ops practices. Excellent communication and people‑management skills, with a passion for mentoring engineers and driving technical excellence. #J-18808-Ljbffr Gravity Engineering Services Pvt Ltd.