Deploy, host, and lifecycle-manage open-source and proprietary large language models on cloud-based container orchestration platforms and on-premises GPU clusters using reproducible infrastructure as code and continuous delivery pipelines
Implement observability across logs, metrics, and traces with dashboards and actionable alerting for large language model and GPU workloads
Tune GPU and accelerator capacity, autoscaling, and cost efficiency for large language model inference workloads using performance optimization techniques such as quantization, parallelism, and speculative decoding
Lead reliability engineering for large language model endpoints through capacity planning, load and soak testing, safe rollouts, failover, and incident response for outages and model-quality regressions
Participate in on-call rotations, lead incident triage and mitigation, and produce clear post-incident root-cause analyses and follow-up actions
Identify recurring operational issues and automate remediation to improve platform stability and developer experience
Build and maintain multi-agent systems with strong orchestration, including planning, coordination, tool-calling, state and memory management, and workflow control where appropriate
Drives team adoption of enterprise-authorized AI-assisted engineering practices within the work environment to improve code quality, delivery speed, and operational outcomes (e.g., AI-assisted code review/refactoring, test strategy acceleration, incident/root-cause analysis support), while establishing consistent validation standards (secure coding, peer review, automated testing) and promoting reuse of effective patterns across the team.
Required qualifications, capabilities, and skills
Hands-on experience with system design, application development, testing, and operational stability in production environments
Advanced proficiency in Python for building production-grade services and tooling
Proficiency with automation and continuous delivery methods
Hands-on experience with cloud infrastructure platforms and infrastructure-as-code tooling for delivery and lifecycle management
Strong understanding of site reliability engineering practices, including incident management, root-cause analysis, runbooks, and reliability patterns
Practical knowledge of observability and instrumentation across metrics, logs, and traces
Hands-on experience with Kubernetes and container-based orchestration platforms, including managed cloud variants
Experience hosting and serving large language models on cloud-based infrastructure and local GPU environments
Knowledge of large language model reliability and risk considerations, including latency and throughput trade-offs, model versioning, prompt and response logging, and safe rollout patterns
Hands-on experience using enterprise-authorized AI-assisted software development tools within the work environment (e.g., for coding, test creation, troubleshooting, or documentation) with demonstrated ability to critically evaluate, validate, and refine AI-generated outputs for correctness, performance, and security
Understanding of responsible AI use in engineering workflows, including data sensitivity considerations, secure handling of inputs/outputs, and adherence to resiliency and security expectations; ability to guide peers on safe and effective usage within team practices
Preferred qualifications, capabilities, and skills
Experience operating large language model inference servers such as vLLM and llm-d (or directly equivalent model serving stacks) in production
Experience developing generative AI applications, AI agents, vector search, and retrieval-augmented generation patterns
Experience building AI agents using orchestration frameworks such as LangChain, LangGraph, CrewAI, or similar platforms
Experience operating or integrating model serving platforms such as KServe, Ray Serve, or NVIDIA Triton Inference Server alongside other large language model serving stacks
Familiarity with Amazon SageMaker JumpStart, SageMaker Endpoints, and Amazon Bedrock for managed model hosting
Experience with online large language model quality monitoring, including hallucination detection, toxicity filtering, and drift detection using open telemetry conventions
Contributions to open-source large language model serving or inference projects, (vLLM, llm-d, Ray, KServe, Triton)