Member of Technical Staff, Applied Research (+ Equity) at well-funded AI infrastructure startup

This role is with a well-funded AI infrastructure startup.

About the role

Join a high-growth AI infrastructure startup building the intelligence layer for complex document understanding. You will develop and train vision-language models to transform unstructured enterprise data into LLM-ready context. This role bridges the gap between frontier research and production systems, allowing you to ship high-impact models used by thousands of developers globally.

Location: San Francisco, USA

Why this role is remarkable

• Work at the intersection of computer vision and language modeling on a core infrastructure problem for the modern AI stack.
• Join a Series A startup backed by top-tier VCs with massive developer adoption and real commercial traction in the enterprise market.
• Enjoy significant technical ownership and the opportunity to work directly with technical founders to shape the future of document intelligence.

Responsibilities

• Develop and train vision-language models specifically optimized for complex document processing, including tables, charts, and multi-page forms.
• Build robust data pipelines for synthetic data generation and create rigorous benchmarking frameworks to evaluate model performance across diverse datasets.
• Collaborate with engineering teams to move successful research prototypes into production-grade systems that handle millions of documents.

Requirements

• 3-7 years of experience in ML engineering or applied research with a strong foundation in training and benchmarking deep learning models.
• Deep technical proficiency in Python and PyTorch, with hands-on experience in computer vision, vision-language models, or natural language processing.
• Proven ability to write clean, production-quality code and thrive in a fast-paced startup environment with high levels of ownership.