NVIDIA’s Worldwide Field Operations (WWFO) team is looking for a Solutions Architect with deep expertise Generative AI post-training from data processing to model alignment, with a strong background in Large Language Models (LLMs). In particular, we are seeking someone skilled in data preparation and curation for large-scale model training, as well as in LLM alignment techniques such as reinforcement learning from human feedback (RLHF) and Supervised Finetuning (SFT).
In this role, you will operate at the intersection of innovative AI research and real-world applications. You will work as the primary technical specialist for NVIDIA customers, helping drive innovation powered by NVIDIA’s hardware and software platform. You will develop proof-of-concept solutions, demonstrate advanced neural network architectures, and help customers adapt and optimize foundation models using NVIDIA technologies for their specific domains. A key part of this role involves close collaboration with a wide range of team members including developers, data scientists, IT managers, and senior executives.
The ideal candidate is an experienced AI specialist with a solid foundation in LLMs and expertise across data processing techniques including large scale data curation, cleaning and Synthetic Data Generation. The candidate should have confirmed experience in Large-Scale post-training using supervised fine-tuning and/or Reinforcement Learning techniques. Hands-on experience applying techniques related to reinforcement learning for LLM alignment is a nice to have.
What You Will Be Doing
* Serve as the primary technical expert to NVIDIA's customers, understanding their technology and providing guidance on post training process in terms of tools and methodology.
* Build proof-of-concepts and demonstrations that highlight the power of NVIDIA AI platforms applied to LLMs post training.
* Partner with developers, researchers, technology specialists, IT professionals, and executives to facilitate the integration of NVIDIA technology.
* Partner with Engineering, Product and Sales teams to develop, plan best suitable solutions for customers. Enable development and growth of product features through customer feedback and proof-of-concept evaluations.
What We Need To See
* MS/PhD or equivalent experience in Computer Science, Data Science, Electrical/Computer Engineering, Physics, Mathematics, other Engineering fields.
* Deep expertise in AI/Deep Learning, with hands-on experience in training and finetuning LLMs.
* Expertise with deep learning frameworks for training LLMs (PyTorch, Megatron-LM, Nemo, Transformers, TRL, VeRL)
* Excellent verbal, written communication, and technical presentation skills in English.
* 6+ years' work or research experience with Python/ C++ / other software development
* AI passionate with a growth mindset, ability to collaborate effectively with different teams (Engineering, Product, Sales, Marketing) in a rapid evolving environment while continuously learning and sharing insights.
Ways To Stand Out From The Crowd
* Familiarity with NVIDIA Nemo Collection (Nemo-Curator, Nemo-RL, Megatron-Bridge).
* Track record in data processing as well as running LLM training and customization with Reinforcement Learning approaches.
* Understanding of HPC systems: data center architecture high speed interconnect InfiniBand, Cluster Storage and Scheduling related design and/or management experience.
Widely considered to be one of the technology world’s most desirable employers, NVIDIA offers highly competitive salaries and a comprehensive benefits package. As you plan your future, see what we can offer www.nvidiabenefits.com/
NVIDIA is committed to fostering a diverse work environment and proud to be an equal opportunity employer. As we highly value diversity in our current and future employees, we do not discriminate (including in our hiring and promotion practices) on the basis of race, religion, color, national origin, gender, gender expression, sexual orientation, age, marital status, veteran status, disability status or any other characteristic protected by law.
JR2008246