Senior Software Engineer - NIM Factory Infrastructure

Senior Software Engineer - NIM Factory Infrastructure page is loaded## Senior Software Engineer - NIM Factory Infrastructure locations:
US, CA, Santa Clara:
US, TX, Remote:
US, NY, Remote:
US, CA, Remote time type:
Full time posted on:
Posted Todayjob requisition :
JR2010744

NVIDIA is the platform upon which every new AI-powered application is built. We are seeking a senior engineer to design and build factory automation for NVIDIA Inference Microservices (NIMs). The right person for this role brings technical drive and creativity to change the way NVIDIA optimizes and serves performant inferencing for every AI model. Our NIM offerings are easy to use, highly performant and tested in all deployment scenarios, in the cloud, on customer’s self-hosted infrastructure and locally on all NVIDIA GPUs.

You will apply your deep technical expertise to design an efficient, scalable and reliable automation factory pipeline that will take AI models to become NIMs that are validated for best in class performance and accuracy.

NVIDIA is building a new category of products, by intersecting our prowess in deep learning and computing, with industry-leading technologies. You will harness groundbreaking technologies, and build a highly efficient factory to power how NVIDIA builds and validates NIMs for inferencing all the way through deployment in heterogeneous hardware and software environments. You will influence and drive technical advances in NVIDIAs workflows and build the infrastructure that strives to accelerate the delivery of every AI model on NVIDIA's GPUs anywhere.

We are looking for technical talent to design and build our factory capabilities, including the underlying infrastructure, pipelines, backends, Docker build, test harness, metrics, performance engineering, log ingestion, and more.
** What you'll be doing:
*** Develop, analyze and optimize factory infrastructure that will take an AI model in and produce a deployable service that is validated across Cloud, On-prem and Kubernetes environments. With the team, define and deliver rapid iterations on the group's technical strategies and roadmaps to deliver and improve the NIM factory. You will be developing harness, automating hardware acceptance, analyze benchmarks, data gathering and statistical analysis of systems health and performance analysis of NIMs
* Work with technical leaders designing and developing scalable and reliable factory acceptance and performance tuning of hardware platforms. You will collaborate with multiple AI model teams to understand their requirements to build an efficient infrastructure that improves every team's productivity.
* Define metrics and drive improvements based on user feedback. You will mentor and collaborate throughout the team and with other teams to grow your colleagues and yourself. You will have a history of learning and growing your skills and those around you.
** What we need to see:
*** A history of using your advanced programming skills to build tooling and automation for hardware system characterization and benchmarking.
* Proven experience debugging and analyzing performance of compute applications and system
* Deep technical expertise working with system software and platform layers including Kernel, device driver, memory, storage, networking and PCIe devices
* Experience working with hardware clusters, distributed system, networking, GPU interconnects (PCie, NVlink), node and cluster interconnect (Infini Band)
* Passion for building platform engineering components and automation of system benchmarking and characterization.
* Excellent interpersonal skills and the ability to lead multi-functional efforts
* BS or MS in Computer Science, Computer Engineering or related field (or equivalent experience)
* 5+ years of shown experience developing performant microservice, cloud software and/or tooling roles
** Ways to stand out from the crowd:
*** Experience delivering optimized system engineering environment for inference applications in data center and consumer grade hardware platforms.
* A history of building and deploying automated benchmarking solution in Cloud and On-prem environments, and their associated CI/CD pipelines
*…