Lead MLOps / AI Platform Engineer

We are seeking a highly skilled Lead MLOps / AI Platform Engineer to design, build, and optimize our next-generation Generative AI and Large Language Model (LLM) infrastructure. This role is pivotal in bridging the gap between cutting-edge AI research and robust production deployment. You will be responsible for orchestrating high-performance GPU environments (specifically leveraging Nvidia H200s), optimizing LLM inference, and maintaining enterprise-grade infrastructure across both Cloud (Google Cloud Platform/Azure) and On-Premise environments., * Deploy, scale, and manage large-scale language models using advanced inference frameworks such as vLLM, TensorRT-LLM, SGLang, and Triton Inference Server.

• Implement and fine-tune performance optimization strategies, including Continuous Batching and advanced Parallelism techniques.
• Conduct load testing, benchmarking, and profiling of LLM deployments using GuideLLM and Locust to ensure optimal latency and throughput.

2. Cloud & Infrastructure Orchestration

• Architect and maintain scalable, secure infrastructure on Google Cloud Platform and Azure using Infrastructure as Code (Terraform).
• Design and execute Cloud Networking, Landing Zones, and Organization Policies/Governance.
• Manage secrets and secure workloads utilizing HashiCorp Vault.
• Develop and champion Internal Developer Portals to streamline workflows for data science and product teams.

3. On-Premise & Kubernetes Engineering

• Orchestrate ML workloads on Kubernetes, utilizing KServe, OpenShift AI / OpenShift Functions, and Helm charts/Operators.
• Manage compute clusters with a heavy focus on advanced GPU Orchestration (Nvidia H200 ecosystems).
• Demonstrate deep hands-on expertise in architecture and "know-how to unfold an LLM" into highly constrained or custom on-premise hardware setups.

4. Observability & SRE

• Implement end-to-end ML Observability and monitoring frameworks using Arize AI.
• Establish Site Reliability Engineering (SRE) best practices, maintaining strict SLOs/SLIs for model deployment pipelines and production APIs.

Core AI / MLOps Stack:

• Inference Engines: vLLM, TensorRT-LLM, Triton Inference Server, SGLang
• ML Frameworks/Ops: KServe, OpenShift AI, Arize AI, GenAI Platforms, RAG architecture
• Performance & Testing: GuideLLM, Locust, Continuous Batching, Parallelism optimization
• Infrastructure & Cloud Stack:
• Cloud Providers: Google Cloud Platform (Google Cloud Platform), Microsoft Azure
• Containerization & Orchestration: Kubernetes, OpenShift, Helm/Operators, GPU Orchestration
• IaC & Automation: Terraform, Python
• Security & Networking: HashiCorp Vault, Landing Zones, Org Policy & Governance
• Hardware Sanity Check:
• Mandatory Experience: Direct, hands-on experience working with Nvidia H200 GPUs and optimizing workloads specifically for this architecture.