Solutions Architect, Open Models

This is a fully remote position, open to applicants in California.

📋 Description

• Assist in technical interactions with customers and partners to comprehend their workloads and determine how NVIDIA’s technology stack can fulfill their requirements, with guidance from senior team members.

• Engage in the design and validation of scalable architectures, integration strategies, and deployment methodologies.

• Construct, test, and validate solutions that facilitate the adoption and scaling of NVIDIA technologies by customers and partners.

• Record and relay customer and partner needs along with technical feedback to internal product and engineering teams.

• Work together with engineering and product teams to ensure successful customer deployments.

• Conduct comprehensive post-training programs with enterprise customers, encompassing supervised fine-tuning, reinforcement learning (GRPO, DPO, PPO), LoRA / PEFT, data curation, synthetic data generation, reward design, and evaluation design.

• Establish evaluation environments to track post-training progress and relay field insights to Product and Research teams.

• Convert practical work into reusable recipes, reference pipelines, and playbooks for NVIDIA teams, customers, and partners.

⛳️ Requirements

• Bachelor’s or Master’s degree, or equivalent experience in Computer Science, Engineering, Mathematics, Physics, Data Science, or a related discipline.

• 2-5+ years of professional experience in software/hardware engineering, developer relations, technical partnerships, solutions architecture, or a similar technical field.

• Practical experience with open-weight model customization and the AI/ML systems ecosystem.

• Fundamental technical understanding of AI/ML, especially in post-training, reinforcement learning, LoRA / PEFT, data curation, and evaluation.

• Capability to articulate technical concepts effectively to both technical and non-technical audiences.

• Strong problem-solving abilities, intellectual curiosity, and a commitment to assisting customers with technology.

• Ability to cultivate relationships with customers or partners and build technical credibility and trust.

• Proficiency in agent-assisted model training workflows.

• Familiarity with AI agents such as Claude Code, Cursor, or Codex for the development, debugging, and iteration of training and evaluation pipelines.

• 3+ years of Linux experience, including executing training and evaluation jobs on SLURM or Kubernetes.

• Experience with and/or contributions to open-source NVIDIA AI libraries and models, particularly Nemotron, NeMo, NeMo-RL, and NeMo Gym.

• Practical experience in designing agent environments, synthetic data pipelines, and verifiable rewards for RL post-training, or utilizing serverless post-training APIs like Tinker.

• Hands-on experience customizing open-weight models such as Nemotron, Qwen, DeepSeek, GLM, and Kimi, or applying AI model training to multimodal data.

🏝️ Benefits

• Equity

• Benefits