Room: TBD
How does research data get made? What research can become "data"? In this track, we'll explore best practices for telling stories and making meaning by collecting, analyzing, and presenting research materials. You’ll learn how to structure and transform images, text, and multimedia resources for use in mapping, text analysis, data visualization, web publication, and digital exhibits. Through shared examples or your own data, this track will also introduce different tools and software platforms available to share, visualize, and publish your research for new and wider audiences. No coding skills are required for this track.
Participants will design and publish interactive maps, multimedia presentations, and digital exhibits to effectively communicate their research findings to broader audiences.
Participants will capture, structure, and transform data and metadata to enable the use of traditional research materials in digital workflows.
Participants will be able to efficiently manage, transform, and back up research files while critically evaluating the accuracy and ethics of generative AI
Participants will practice reusing data from APIs in a web context.
Day 1: Foundation & Exploration – Finding, collecting, and structuring data
Participants will explore the use of text as data, including collecting, organizing, and transforming public domain OCR plain text files
Participants will learn how to find data from public sources, specifically NYC Open Data.
Participants will practice capturing and structuring metadata.
Participants will gain an understanding of API and how to use cultural heritage materials using IIIF APIs.
Participants will implement research data management and documentation best practices
Day 2: Mapping & Visualizing – Working with images and geospatial data
Participants will learn about different types of geospatial data (raster vs. vectors)
Participants will learn how to georeference images of maps onto geospatial coordinates using AllMaps.
Participants will exhibit images and geospatial data using IIIF tools and Esri StoryMaps.
Day 3: Synthesis & Digital Storytelling – Exhibiting data to tell a story
Participants will work alone or collaborate in groups to continue developing a proof-of-concept for a data storytelling project based on the tools and data introduced on the previous days.
Room: TBD
Generative AI offers an exciting opportunity to interact with large amounts of data and discover connections in ways that were not previously possible with traditional research methods. However, commercial Generative AI products offer a “black box” for users, who may not know what Large Language Model (LLM) is being used or how queries are being modulated, all factors that can affect the quality and usefulness of outputs. In this track, you will learn about differences in LLM offerings and their applications, standard methods for modulating outputs, and basic contours for narrowing and improving outputs using a retrieval-augmented generation (RAG) workflow. Students will learn how to configure NotebookLM to meet their research needs. No coding skills are required.
Confidently deploy NotebookLM and Gemini to strategically delegate tasks to AI based on their personal mastery and domain expertise.
Utilize advanced prompting frameworks to communicate precise research goals and constraints to Large Language Models.
Critically evaluate AI outputs to identify issues such as "flattened" nuance, inaccuracies, and biases.
Implement scholarly transparency by documenting AI-assisted processes and manually verifying generated citations against primary sources.
Day 1: Generative AI Foundations - NYU Tools, Prompting, AI Fluency Framework
Day 2: Evaluating Generative AI Outputs - Delegation, Description and Gemini Gems
Day 3: AI Research Assistants in Action - Deploying Gems
Room: TBD
In this track, participants will get an understanding of how to find, understand, and manage data as well as hands-on experience with using Python for computational research purposes. We’ll explore different libraries commonly used for data analysis and finish strong with creating visualizations in Jupyter notebooks. No prior coding knowledge needed.
Gain knowledge and build confidence about the computational research journey
Learn how to setup your computational work environment
Understand core principals and syntax of Python
Gain facility with common Python libraries and research applications
Create data visualizations with Python
Day 1 - Intro to Research with Data
Overview of the Research Data Lifecycle - what to expect + resources available to you
Introduction to research data management and the command line
Setting up your computational work environment
Day 2 - Getting Hands-On with Python
Understand core Python classes used all the time in writing code (strings, integers, lists, dictionaries, Booleans)
Be able to compose additional helpful structures, namely loops and functions
Use Python to open two types of common research data files (text and table/csv) and work with Pandas DataFrames
Perform a few example research applications with data files such as modifying text and tables
Day 3 - Showcasing Data Visualization
Introducing commonly used data visualization libraries in Python
Fine-tuning our data visualization skills
Hands-on practice along with a prompt for our final showcase
Room: TBD
Join our multi-day R workshop designed specifically for graduate students. This series will teach you the modern, highly efficient tools needed to analyze real-world datasets with confidence, while introducing Generative AI as an assistant to accelerate your workflow.
We will introduce RStudio environment and foundations of R language, explore core tidyverse packages for importing, cleaning, wrangling, and ultimately visualizing datasets. We will also explore AI assisted coding. By the end of the series, students will have explored a dataset, investigated relationships within, and presented their findings to their peers.
No prior coding knowledge needed.
This series will teach students the modern, efficient tools needed to analyze real-world datasets with confidence, while introducing Generative AI as an assistant to accelerate your workflow.
We will introduce RStudio environment and foundations of R language, explore core tidyverse packages for importing, cleaning, wrangling, and ultimately visualizing datasets. We will also explore AI assisted coding. By the end of the series, students will have explored a dataset, investigated relationships within, and presented their findings to their peers.
Day 1: Foundation & Data Wrangling
R Foundations: Master the RStudio environment, essential data structures, and core coding mechanics.
Data Wrangling: Use the tidyverse packages to effortlessly clean, filter, and reshape messy datasets.
Day 2: Data Visualization & AI
Data Visualization: Learn ggplot2 to transform raw numbers into compelling, customizable graphics.
AI-Assisted Coding: Learn about GenAI strengths and limitations in the realm of data analysis and explore caveats and pitfalls for various stages of data analysis (EDA, data cleaning, statistical analysis, and data visualization).
Day 3: Synthesis & Storytelling
The Capstone Showcase: The workshop culminates in a hands-on "Data Briefing" showcase. You will apply your new R skills to a real-world dataset—cleaning the data, generating a professional-grade visualization, and sharing your findings with the group.
No prior coding experience required. Please install R and RStudio prior to the first session.
Room: TBD
In this track, users will learn about programmatic use of LLMs beyond the chatbot. The first day will cover the basics of LLM APIs, parameters and tools. The second day of the workshop focuses on Retrieval Augmented Generation (RAG) and how it enhances AI models by combining external data retrieval with large language models (LLMs). They will explore the steps to build a RAG pipeline, including embedding text into vector representations, retrieving relevant context from databases, and augmenting prompts to generate accurate answers. The tutorial provides practical insights into when to use RAG over fine-tuning models and how to integrate this approach in building dynamic, context-aware AI solutions.
For additional information, see Retrieval Augmented Generation (RAG) page.
Prerequisites: basic understanding of Python