2026 - 2027 SURE Projects

Project Number

Project Title

Brief Project Description

Job Duties

Required/Desired Skills

Project # 1

Project Title:

Building out an AI enabled search and summary tool for historic Wyoming newspapers

Description:

Over the past year a group of undergraduate students have started to create an LLM driven Optical Character Recognition (OCR) pipeline to pull out individual newspaper articles from digitized media. This pipeline was built using Python. Our next steps are implement an MCP Vector database approach to allow for discovery of articles following a specific topic. Secondly, we would like to build a web interface to allow the public to access these historic articles.

Job Duties:

- Write python to interact with LLM
- Work with team members on management of the Postgres Vector database
- Build web API and interface to enable article discovery

Required Skills:
- Comfortable in writing code, specifically Python.
- Familiar with code repositories
Desired Skills
- Familiar with C#, HTML, JavaScript, CSS
- Familiar with fine tuning of LLM workflows, vector databases, SQL

Project # 2

Project Title:

Investigating the movement and survival of juvenile golden eagles

Description:

SURE undergraduates will have the opportunity to study the early-life movements and survival of golden eagles in the western USA, using multi-sensor datastreams (i.e., GPS, accelerometry and possibly IMU). During the first years of life, golden eagles are highly mobile, often moving long distances into unfamiliarly terrain that exposes them to novel risks. Depending on student interests, they could work on the following projects: 1) characterizing the range and diversity of exploratory movements in golden eagles, 2) testing, documenting and enhancing existing methods to identify novel explorations, with the goal of building a general tool that can be used by the research community, 3) building a new tool to enhance survival monitoring of golden eagles that combines both GPS and accelerometry data. All of these projects could result in a peer-reviewed publication, if the student/students has interest in pursuing this.   

Job Duties:

Analyze GPS and accelerometry data, write new code using reproducible data science practices, document research finding, participate in or lead the development of a peer-reviewed publication 

Required Skills: 

Previous experience coding in python, previous experience using version control (github)
Preferred: familiar with geospatial data, an interest in wildlife or raptor ecology

Project # 3

Project Title:

Technology Infusion and AI Literacy in Rural Classroom Practice

Description:

This project focuses on strengthening technology infusion and AI literacy across teacher preparation by making expectations for purposeful technology use more visible, consistent, and actionable across coursework and clinical experiences. The big idea is to better understand how teacher candidates and teacher educators currently define, prioritize, and experience technology infusion and AI literacy, then use those insights to create shared tools that support more coherent preparation, observation, feedback, and reflection.

The SURE student will work on part of the project that includes a co-creation process for an AI self-evaluation tool for users' integration of classroom technology. We will work with teacher candidates to co-create an AI dashboard self-evaluation tool that helps them interpret feedback, reflect on their growth, and evaluate their developing technology infusion and AI literacy practices. Rather than designing the dashboard for candidates, this part of the project centers candidates as design partners who help determine what feedback is most useful, which data to include, and how progress should be represented visually.

Job Duties:

The SURE student will assist with the iterative prototype process of the AI tool. Additionally, the SURE student may help with the user feedback collection: organizing student feedback sessions, documenting user needs, collecting and summarizing feedback from teacher candidates and other stakeholders, and supporting revisions to the dashboard prototype across multiple design cycles. This feedback will be used to iterate the prototype. Through this work, the SURE student would contribute to the development of a usable, student-informed tool that supports reflection, agency, and growth by helping candidates identify strengths, recognize patterns in their practice, and determine next steps for improving technology infusion and AI literacy.

Required Skills:

Basic coding or technical troubleshooting skills
Organization and communication skills
User feedback collection skills
Interest in AI literacy and educational technology
Dashboard, visualization, or design tool experience a plus

Project # 4

Project Title:

The EDGES (Ecological Directed Graphs for Evidence Synthesis) Project: Using AI and Data Science to Improve Causal Inference in Ecology

Description:

How do ecologists determine what actually causes changes in nature when controlled experiments are often impossible? This is one of the greatest challenges in field-based sciences such as conservation biology, population biology, and behavioral ecology. While ecologists regularly study systems shaped by predators, competition, habitat loss, biodiversity, and human impacts, many—if not most—scientific studies still struggle to separate true causal relationships from simple correlations.

The EDGES Project in the Behavioral Complexity Lab, Department of Zoology and Physiology, combines ecology, classical metrology, data science, and artificial intelligence to improve how ecological science is conducted and interpreted. Undergraduate researchers who join this project will use AI-assisted textual analysis, scientific literature mining, and reproducible workflows in R and Python to analyze tens of thousands of published ecology papers and quantify how scientists make causal claims.

Students in the EDGES Project will gain experience with data science, visualization, and modern approaches to causal reasoning using Structural Causal Models (SCMs) while contributing to the BCL’s active research program focused on improving scientific inference in ecology. The project will also help prepare students for future BCL opportunities in tropical field ecology and causal inference research in rainforest systems in Panama and Hawai‘i.

Job Duties:

Assist with collection and organization of scientific literature datasets
Use R and Python workflows to clean, summarize, and visualize data
Help annotate ecological papers for causal, associational, and descriptive language
Assist with AI-assisted text analysis and natural language processing workflows
Create figures, summary graphics, and exploratory visualizations
Participate in collaborative lab meetings and project discussions
Help prepare presentations and materials for Undergraduate Research and Inquiry Day

Required Skills:

Interest in ecology (including behavioral ecology), environmental science, data science, AI, statistics, or scientific reasoning
Willingness to learn computational research workflows (and have fun doing it!)
Strong attention to detail and curiosity about really big and important scientific questions
Prior coding experience in R or Python is helpful but definitely not required
Students from a wide range of majors and experience levels are encouraged to apply!

Project # 5

Project Title:

TRACE: Testing, Reliability, Adaptive Calibration, and Evaluation of Ecological Sensors

Description:

How do scientists reliably measure animal behavior and environmental change in difficult real-world environments? Modern ecology increasingly relies on low-cost acoustic and environmental sensors to study wildlife, yet heat, humidity, rainfall, power limitations, and sensor drift can all affect scientific measurements.

The TRACE Project in the Behavioral Complexity Lab (Department of Zoology and Physiology) combines ecology, acoustics, environmental sensing, and data science to develop and test modular ecological sensing systems for wildlife research. Undergraduate researchers will help assemble, test, calibrate, and analyze low-cost sensor systems designed for rainforest and field environments. Students may contribute to projects involving: (1) building modular acoustic and environmental sensors, (2) stress-testing sensors under humidity and heat, (3) analyzing 3D acoustic localization datasets, (4) visualizing animal movement and sound localization, (5) testing sensor reliability and timing accuracy,
and developing reproducible workflows in R and Python.

Students will gain experience with ecological sensing systems, environmental data analysis, acoustics, visualization, and scientific computing while contributing to active research in tropical ecology and wildlife monitoring.

Job Duties:

Assist with (and sometimes work independently on) construction and testing of ecological sensing systems
Help collect, organize, and visualize environmental and acoustic datasets
Use R and Python workflows to analyze sensor performance and uncertainty
Conduct environmental stress tests involving humidity, heat, and rainfall simulations
Assist with AI-assisted detection of sensor anomalies or failures
Create figures, dashboards using R or other tools, and summary visualizations
Participate in collaborative lab meetings and project discussions
Help prepare presentations and materials for Undergraduate Research and Inquiry Day

Required Skills:

Interest in the nuances of ecology, environmental science, engineering, AI, acoustics, electronics, or data science
Willingness to learn computational and sensor-based research workflows
Desire to imagine, create, test, break, and improve new methods and ideas!
Strong attention to detail and extreme curiosity
Prior coding or electronics experience is helpful but not required
Experience with R, Python, Arduino/Teensy systems, spreadsheets, or data visualization is desirable but not necessary
Students from a wide range of majors and experience levels are encouraged to apply.

Project # 6

Project Title:

Using artificial intelligence for physical education teacher wellbeing support

Description:

My research team and I have studied the psychological resources (i.e., adaptability, coping, emotion regulation, etc.) that physical education teachers needs to be highly motivated in their jobs, which means offering high quality education experiences for increased student outcomes. However, the demands of this job often reduce PE teacher wellbeing and motivation leading to inappropriate or poor practices, which severely  undermines student learning in K-12 PE in schools. Our research team is aiming to develop a personalized phone application that is AI-based that offers teachers individualized support, feedback, and connections for continual professional support. We will used psychometrics to develop psychological wellbeing profiles of teachers, and then use AI to interpret their greatest individual needs to then write code that leads to tasks, suggestions, and practical steps to support teachers. All of this would be collected on a cell phone-based app for teachers.

Job Duties:

We need someone who knows how to write code to create a platform for the app. We know the content and functions, but our research team has no computer science or technology background in terms of developing this tool. We essentially need someone to build the phone application for us while we offer what the functions need to be, and how AI will be incorporated to assist with real time decisions based on the information we provide it. We would have monthly meetings and phases of development to build the app with deadlines, next steps, etc.

Required Skills:

Skills to build a phone-based application that includes many functions (menus, categories, other teaching-based supports, and AI incorporation for data collection and support suggestions to the user).

Project # 7

Project Title:

Topological Neural Networks for Materials Science

Description:

Graph neural networks (GNNs) are machine learning models that encode structured data into graphs made of nodes and edges. In chemistry and materials science, GNNs are particularly powerful as they can model how atoms and molecules are connected, allowing researchers to rapidly predict the structure and properties of materials. However, standard graphs may not capture the more complex geometric relationships within materials which often influence properties like electrical or thermal conductivity. When these complex geometric relationships remain unchanged under certain transformations, they are called topological relationships. This project aims to develop a Topological Neural Network (TNN) by expanding GNNs to include the additional topological information. The project will focus on Metal-Organic Frameworks (MOFs), a class of materials built from molecular building blocks with well-defined geometries. This produces materials with well characterized topologies that are strongly linked to their bulk properties, thus providing an ideal system for developing and testing a TNN for materials science.

Job Duties:

We envision 2-3 students working on this project. One or two students would focus on the expansion of a GNN into a TNN while another student would focus on accessing the Quantum MOF (QMOF) database and processing the material data. For task 1, students would develop and test code, primarily within Python; perform topological data analysis; and test and optimize neural network algorithms. For task 2, the student will interact with a cloud database; process material structure data, including topological data analysis; and simulate material structure and properties with quantum mechanical software.

Required Skills:

Minimum qualifications include Python proficiency; fundamental programming knowledge equivalent to an introductory programming course; time management; and communication, especially within a team. Desired qualifications include familiarity with crystallography; graph networks; mathematical topology; machine learning; and utilizing high-performance computing infrastructure.

Project # 8

Project Title:

Comparative Model Evaluation for OCR pipelines on Mass Market Paperbacks

Description:

This project addresses a significant and rapidly-changing challenge in computational text analysis: corpus creation and digitization. Instead of relying on a single, commercial solution for optical character recognition (OCR), we will build a comparative pipeline that tests multiple state-of-the-art OCR models and deep learning techniques. We will systematically evaluate model performance across varying sources (e.g. Does 80s Star Trek present different digitization challenges than contemporary Halo novels?), layouts, and paper quality, and test when human-in-the-loop evaluation is most impactful. If you want to understand the complete digitization pipeline for text (the foundation for LLMs) while reading a wide range of science fiction and fantasy, this is the project for you!

This work contributes directly to a digital scholarly corpus, and student authorship of that corpus will be attributed in persistent metadata. This work will also contribute to a methods paper; all students who participate will be co-authors.  

Job Duties:

Data curation: destructive scanning of novels, manual annotation, validation, and gold-standard data creation of small batches of source material for training and testing models.
Model testing and benchmarking: running comparative tests on multiple OCR models and pipelines
Performance analysis: developing scripts to calculate and analyze key performance indicators, such as character error rate and word accuracy
Documentation: participating in lab-wide documentation to support the above, up to and including learning to write scholarly digital humanities work.

Required Skills:

Minimum qualification: interest in digital humanities, literature, or digitization. Detail-oriented, comfortable (not expert) with spreadsheets.
Desired qualifications: Proficiency in Python, familiarity with machine learning frameworks
Bonus qualifications: fluency with HPC.

Project # 9

Project Title:

Grief Online: Using Digital Humanities to Understand Loss

Description:

Job Duties:

Since students will be accessing this data most likely through an application programming interface (API), daily tasks include code development and testing. Statistical analyses of the results will likely include, but not be limited to, word frequency, pattern recognition, clustering, and sentiment analysis. Students will also be expected to meet with the overseeing faculty on a weekly or bi-weekly basis to update on progress and ask for guidance when necessary.

Required Skills:

The only requirement is a willingness to learn, problem-solve, and work collaboratively and effectively within a team. Familiarity with programming fundamentals and/or experience with programming languages, like R or Python, will be helpful, but not required. Likewise, familiarity with APIs; statistical analysis; data curation and management; utilizing structured query language (SQL) databases; manipulating JSON files; and experience with human subjects research are all helpful, but are not required.

Project # 10

Project Title:

Agent-Based Modeling for Natural Hazards Adaptation: Modeling how people and policymakers prepare for and respond to natural disasters

Description:

This project aims to support tourism-dependent communities as they adapt to a future with more wildfire. We will model decision-making in the face of wildfire, supporting a large, interdisciplinary project exploring the interactions of atmospheric science, forest fire, and socio-economic systems in gateway communities. Specifically, the student will build an agent-based model that simulates land use and land management decisions by large landowners, public land managers, and city planners. In later work, which the student will be welcome to participate in after the conclusion of the SURE program, this model will be linked to forest fire and econometric simulations driven by a high-resolution atmospheric science model, allowing the team to simulate fire regimes and socio-economic outcomes under a variety of possible environmental conditions and policy agendas. In this scheme, the agent-based model serves a vital linking role, tying observed fire outcomes in forests to decisions that may impact both recreation outcomes and future fire events. Under Jake’s guidance, the student will contributeto building the “missing link” for understanding wildfire futures in Wyoming. Students should join this project if they are interested in environmental systems, interdisciplinary computing, human decision-making, wildfire, tourism, or policy. 

Job Duties:

Interpretation of cognitive maps and other decision-making interview results; Implementation of decision-making rules in an agent-based model; Model interpretation, refinement, and optimization for fit against historic land use choices.

Required Skills:

Knowledge of Python preferred but not required; Interest in human-environmental systems and natural hazards (e.g., wildfire, landslide, flooding) preferred.

Project # 11

Project Title:

Modeling Gentrification in Small Towns: Do mountains and lakes make homes unaffordable?

Description:

Housing is unaffordable almost everywhere. Housing is well beyond unaffordable, and approaching catastrophically out of reach, in communities with natural amenities (e.g., mountains, lakes, trails). Why is this? Are certain types of amenities more prone to this? Do patterns look different in different states? To answer these questions, we need to build the first nationwide dataset of gentrification (i.e., community displacement via housing prices). Building on a recent project exploring gentrification along Michigan’s Great Lakes shorelines, this work will expand the analysis to the Intermountain West and beyond. We will use a novel method developed by the Human-Environmental Systems Lab to translate big data into big insights about housing price evolution in amenity-rich communities. Students should join this project if they’re interested in housing, human-environmental systems, tourism, recreation, or big data.

Job Duties:

The student will build on an existing code base which conducts the big data analysis for Michigan. They will be responsible for building a gentrification database for Wyoming and ensuring the code is generalizable to at least one other state. If the student is interested in High-Performance Computing, students can be taught how to run this code on ARCC or NWSC and conduct the analysis for the entire Intermountain West or nationwide.

Required Skills:

Knowledge of Python or R preferred but not required; Interest in human-environmental systems or public policy or big data preferred.

Project # 12

Project Title:

Wyoming Soundscapes: Passive acoustic monitoring to detect species and community change

Description:

Wyoming’s wildlife communities face accelerating pressures from climate change, energy development, and land-use transformation, yet robust, scalable tools for tracking biodiversity across its varied landscapes remain limited. This Wyoming Soundscapes project aims to deploy passive acoustic monitoring (PAM) in two ecologically distinct habitat types: the sagebrush steppe and the montane forest.
The intention is to test whether soundscape features can predict species occurrence and community change in Wyoming ecosystems. Building on machine-learning frameworks of Sethi et al. (2022, Oikos), BirdNET, and Perch 2.0, this project takes a cost-effective, technology-forward step toward autonomous, long-term wildlife monitoring scalable to landscapes statewide.

Job Duties:

Acoustic data processing and wrangling; Python coding (machine learning, processing scripts); field recorder deployment; data annotation; writing; critical thinking.

Required Skills:

Python proficiency, especially with Pandas and scripting for handling large amounts of file I/O.
Terminal user / linux proficiency.
Proficiency with common machine learning libraries like PyTorch and scikit-learn is a huge plus.
Experience with acoustic libraries like librosa, OpenSoundscape, etc. is an even bigger plus.
Interested in wildlife conservation.

Project # 13

Project Title:

AI-Enabled Mobility Knowledge Extraction for Rural Public Health Modeling

Description:

Wyoming transportation corridor planning documents contain rich information on mobility patterns, freight movement, tourism, weather disruptions, and infrastructure needs — but it is locked in long PDFs. This project uses AI and natural language processing to transform those documents into structured, searchable datasets that can support future research on rural mobility systems and public health modeling.
The student will build reproducible Python workflows for extracting, organizing, and analyzing transportation reports from Wyoming State Significant Corridor (SSC) planning documents.

Job Duties:

Extract text and tables from transportation planning PDFs
Organize corridor metadata and regional transportation information
Identify key transportation and mobility themes across documents
Test AI-assisted extraction approaches using LLMs and NLP tools
Build structured datasets from semi-structured reports
Develop reproducible Python pipelines for document processing
Participate in weekly meetings and present research outcomes

Required Skills:

Python programming 

• NLP / text processing 
• PDF / document parsing

Desired Skills:

• AI / LLM tools (OpenAI API or open-weight models) 
• Geospatial / GIS 
• Semantic search (FAISS, sentence-transformers) 
• Data engineering 
• Interest in transportation or public health.

Project # 14

Project Title:

AI-Enabled Socioeconomic Knowledge Extraction for Rural Systems Analysis

Description:

Wyoming regional studies and socioeconomic reports contain detailed information on demographic change, economic transition, tourism, labor conditions, housing pressures, and rural development — all factors closely linked to how people move across communities. This project applies AI and NLP to convert those lengthy reports into structured socioeconomic datasets.
The student will develop Python-based workflows for extracting and organizing information from reports on tourism, workforce conditions, community development, and regional indicators.

Job Duties:

Extract text and tables from regional socioeconomic PDFs
Organize study metadata and regional socioeconomic indicators
Identify major socioeconomic and community development themes
Test AI-assisted extraction approaches using LLMs and NLP tools
Build structured datasets from semi-structured regional reports
Develop reproducible Python pipelines for document processing
• Participate in weekly meetings and present research outcomes

Required Skills:

• Python programming 
• NLP / text processing 
• PDF / document parsing
  

Desired Skills:

• AI / LLM tools (OpenAI API or open-weight models) 
• Structured data organization  ·  Semantic search (FAISS, sentence-transformers) 
• Data visualization 
• Interest in regional analysis or social scienc.

Project # 15

Project Title:Human settlement damages dataset for training Deep Learning networks

Description:

Several pre-trained AI/Deep Learning (DL) neural network models are available for identifying buildings, roof-tops, and infrastructure (road intersections, road center lines, etc.) in aerial photos and are integrated in many geospatial software.  However, DL models for identifying damaged buildings and infrastructure (following tornadoes, floods, and wildfires) are not that common. Trained DL models that can identify damaged buildings and infrastructure are needed for rapid disaster response.

Training AI/Deep Learning models require 1000s of post-disaster images depicting different damage types (missing roof, damaged roof, broken fence, dock damage, etc. etc.).  Currently such datasets are not readily available in public domain. Imagery datasets depicting various damage types are in need for training DL networks.

The proposed project will build a dataset consisting 1000s of aerial images depicting 10 distinct types of damages to human settlements. High-resolution aerial images acquired after big and small disasters will be used to generate this dataset, which will be published in public domain (e.g., Kaggle) along with metadata.

Job Duties:

Download post-disaster aerial images from USGS and other US government agency websites, interpret damages (1 of 10 types), generate damage dataset, write metadata files, and upload them to public sites like Kaggle.

Required Skills:

Aerial image/photo interpretation, familiar with remote sensing, digital file organization skills.

Project # 16

Project Title: Identifying Spatially Informative Genes in the C. elegans Epidermis

Description:
The C. elegans epidermis contains a large multinucleate cell called hyp7, in which many nuclei with different developmental origins share the same cytoplasm. It is unknown whether these nuclei become transcriptionally identical or retain distinct regional or lineage-specific identities. This project will integrate single-cell, single-nucleus, developmental, and lineage-resolved gene-expression datasets to identify genes that distinguish subsets of hyp7 nuclei. The goal is to generate a ranked set of candidate spatial markers that can be experimentally validated and used to map nuclear specialization within the syncytium.

Job Duties:

• Analyze existing single-cell and single-nucleus gene-expression datasets
• Identify genes that distinguish hyp7 nuclear populations or developmental lineages
• Develop computational methods to rank genes by their ability to predict nuclear identity
• Build visualizations and a searchable resource of candidate hyp7 marker genes

Required/Desired Skills:
Interest in biology, genetics, data science, or bioinformatics. Experience with Python or R is helpful but not required. Familiarity with statistics, machine learning, or genomic data analysis is desirable.

Project # 17

Project Title: Machine learning driven modeling of social salmon migration

Description:

This project is focused on studying the social dynamics of salmon migration. Social migration has interesting ecological implications while also being an ideal natural system for exploring ideas related to network dynamics and complex systems. The core focus of this project consists of designing mathematical models that can be paired with real data derived from continuously recording security cameras deployed along salmon migration streams. These cameras allow us to watch the patterns of salmon arrival across a network of nesting ponds. The video is processed with computer vision models that can automatically process 10s of millions of video frames. The mathematical models are used to understand the relationship between the natural patterns we see and the possible underlying mechanisms that might generate such patterns. 

Job Duties:

Write python code that can be run on the AI4WY cluster to help process the salmon videos. Specify and implement mathematical models and agent based simulations (also in python) that model various aspects of this natural system.

Required/Desired Skills:

Comfort designing mathematical models; experience programming in python; interest in ecological processes.