NYBG Job Description

Title

Reports to

Postdoctoral Research Associate

Director of the Herbarium

Position Summary:

A postdoctoral research associate is sought to develop tools to mobilize data in herbarium specimens for conservation action, as part of a project funded through the Bezos Earth Fund’s AI for Climate and Nature Grand Challenge. The postdoctoral research associate will report to Emily Sessa, Patricia K. Holmgren Director of the William and Lynda Steere Herbarium at NYBG.

Herbaria—scientific collections of preserved plant specimens—represent one of the most important frontiers for plant species discovery and for understanding, assessing, and protecting threatened biodiversity. Globally, billions of herbarium specimens contain critical information on species distributions, traits, and environmental responses, yet much of this information remains locked in images and unstructured labels. This project harnesses recent advances in artificial intelligence to unlock these underutilized data at unprecedented scale, with the goal of accelerating biodiversity discovery, informing conservation priorities, and supporting evidence-based environmental decision-making. The postdoctoral research associate will play a central role in this effort by designing and implementing a web-based AI tool that takes digitized herbarium specimen images as input and produces structured textual outputs, including specimen-level data extraction and synthesized summaries, specifically tailored to support IUCN Red List assessments for species of priority conservation concern. Responsibilities will include developing and integrating vision-language and computer vision models, translating AI outputs into formats aligned with Red List criteria, and collaborating with conservation scientists to ensure the tool is accurate, interpretable, and usable for real-world conservation assessments.

Specific Duties & Responsibilities:

●       Design, develop, and implement a web-based AI tool that ingests digitized herbarium specimen images and outputs structured textual data to support IUCN Red List assessments.

●       Develop, train, and evaluate computer vision and vision-language models for tasks such as specimen interpretation, trait extraction, label parsing, and metadata synthesis.

●       Translate model outputs into structured specimen-level data and higher-level summaries aligned with IUCN Red List criteria (e.g., distribution, population trends, threats, and habitat information).

●       Collaborate with AI and conservation partners to define requirements, validate outputs, and ensure biological and conservation relevance.

●       Design workflows to handle large-scale herbarium image datasets, including data preprocessing, quality control, and performance evaluation.

●       Implement and maintain reproducible, well-documented codebases, following open science and best practices in research software development.

●       Contribute to the deployment and maintenance of AI models and web services, ensuring usability, scalability, and accessibility for end users.

●       Assess and document uncertainty, bias, and limitations of AI-generated outputs, particularly in conservation decision-making contexts.

●       Prepare manuscripts, technical documentation, and/or project reports for peer-reviewed publication and stakeholder dissemination.

●       Engage in a collaborative postdoctoral mentoring plan.

●       Participate in NYBG Science activities and outreach related to the project.

Qualifications:

• PhD in a relevant field (e.g., computer science, data science, bioinformatics, biodiversity informatics, ecology, botany, evolutionary biology, or a related discipline).
• Demonstrated experience with computer vision and/or vision–language models, particularly for image analysis or multimodal data.
• Proficiency in Python and common ML frameworks (e.g., PyTorch, TensorFlow, JAX), as well as experience working with