About the job
As a Senior / Principal Deep Learning Engineer — Model Evaluation & AI Systems, you will play a meaningful role in crafting the future of AI. Your work will have a direct impact on our product releases and positioning in the market.
Responsibilities
Define and build evaluation methodologies for innovative AI models, including LLMs, RAG systems, agents, and vision/multimodal models.
Build and expand NeMo Evaluator as an open-source platform, focusing on correctness, reproducibility, and ease of adoption.
Build scalable, reproducible evaluation infrastructure, including harnesses, orchestration, and result pipelines running on large GPU clusters.
Collaborate with and engage the open-source community, reviewing contributions, shaping the roadmap, and sharing best practices.
Work alongside model training, inference, and product divisions to provide trusted evaluation signals that inform release and optimization decisions.
Qualifications
Minimum
BS, MS, or PhD in Computer Science, AI, Applied Math, or a related field, or equivalent experience.
Senior-level experience (typically 12+ years) developing or assessing contemporary machine learning and deep learning systems.
Hands-on experience with large language models and NLP, including model behavior analysis and evaluation.
Demonstrated experience contributing to open-source software or building platforms, libraries, or tools used by other engineers.
Ability to take charge of unclear technical challenges and communicate effectively across research, engineering, and product teams.
Preferred
Experience building or improving evaluation frameworks, benchmarks, or ML infrastructure used by other teams or external users.
A strong appreciation for evaluation quality, including correctness, reproducibility, and consistency across environments.
Hands-on experience evaluating modern AI systems such as LLMs, RAG pipelines, agents, or multimodal models.
Prior involvement in open-source projects, through contributions, reviews, maintenance, or community engagement.
Experience acting as a technical bridge across teams or platforms (e.g., evaluation, training, or agent frameworks), combining architectural understanding with clear communication and influence.