Past Summer Projects
CICF Student Fellows have the option to apply for a hands-on project-based learning experience at either an MF or a CI Compass institution during the summer. Spaces are very limited, so summer placement is not guaranteed and requires an additional application process that takes plce during the Spring Program.
Here is a list of past summer projects:
Summer 2025
This summer, I worked with Ocean Observatories Initiative (OOI), where I investigated ocean temperature variability and its connection to El Niño and La Niña events. Using Python in a Jupyter Notebook environment, I integrated long-term datasets from OOI, Ocean Networks Canada (ONC), and the National Ecological Observatory Network (NEON), where I employed multiple data retrieval methods, including ERDDAP, NEON’s Data API, and direct file downloads to compile a consistent time series of data. Ensuring thorough documentation, I compared and analyzed seawater temperature and air temperature across the East-Central Tropical Pacific region whilst correlating them with El Niño and La Niña events. Through this project, it has deepened my understanding of differences in data infrastructures, particularly in instrumentation granularity, data accessibility, and sensitivity to ENSO signals across regions.
This summer, I had the opportunity to work at the National Center for Atmospheric Research (NCAR). Throughout my time there, I contributed to the CESM (Community Earth System Model) Unified Postprocessing and Diagnostics (CUPiD) package. I developed and modified Python scripts to process and visualize large-scale earth system model outputs using high-performance computing resources. My work involved generating diagnostics for high-resolution CESM simulations, exploring methods for remapping data to different grids, and improving workflow automation. Through collaboration with scientists and software engineers, I gained valuable experience in scientific computing, climate data analysis, and the development of tools to support ongoing research.
Raymond James Gallant - STEMSEAS (CICF 2024)
Over the past year, I’ve completed three cybersecurity internships that shaped my perspective on what it means to grow in this field. I started in summer 2024 at the NSF National Center for Atmospheric Research, where I worked on high performance computing and parallel workflows with the CESM team. I came in with no programming background and felt completely out of place, but my mentor helped me learn on the job and build a technical foundation from scratch. In spring 2025, I joined the UNOLS-MATE SECURE-IT Cyber Training Cruise aboard the R/V Sikuliaq. I deployed Fortinet firewalls, worked with OpenRVDAS, and saw firsthand how scientific infrastructure is secured in harsh, real-world environments. Across these experiences, I’ve learned that success in cybersecurity isn’t about being the most technical person in the room. It’s about staying adaptable, asking questions, and learning through discomfort.
During my time at the Ocean Observatories Initiative, my project focused on applying convolutional neural networks (CNNs) to detect and classify ice floes using digital stills captured by a surface buoy at the Global Irminger Sea Array. To process the raw image data stream, I implemented unsupervised pretrained models and Vision Transformers to isolate specific features. In the latter half of the project, I extended the pipeline to classify cloud formations using the same image dataset. I utilized the PyTorch framework for model development and DINOv2 and FAISS for image embedding and clustering during the initial processing stage.
During the summer I had the pleasure of working at NCAR, working on their global fish population model, FEISTY (A Fisheries Size and Functional Type Model). During my time, I debugged a glitch in switching between an old and a new python environment with updated versions and new libraries. In my work, I delved deep into learning the structure of the model, the importance of various variables, and the relevance and accuracy of produced results. My biggest breakthrough was in comparing generated metadata from running both environments, and seeing an inconsistency between how data was temporarily chunked. An update in the Xarray library caused this, and my fix was to force the model to chunk in a specific way no matter the version. This solved most issues, and results produced were identical. This experience was critical in expanding my knowledge in python, jupyter notebooks, and debugging.
Naomi Kolodisner - Globus/UChi
This summer, I was an intern with Globus Labs at the University of Chicago, where I worked on converting a bioinformatics pipeline into an agent-based scientific workflow using the Academy framework. My project focused on transforming a viral detection genomics workflow, originally implemented as a linear sequence of Slurm batch scripts, into a modular, adaptive system composed of autonomous agents. The original workflow processes metagenomic samples to identify viral sequences, assess quality, dereplicate and cluster them, and annotate them using various bioinformatics tools. I reworked the stages of my workflow into discrete, parallelizable agents and integrated decision logic that dynamically selects the best viral detection tool based on performance metrics. Along the way, I also explored parallel task execution with Parsl and ran scaling experiments on an HPC cluster to optimize resource use. This internship was an incredibly rewarding experience that challenged me to grow and learn in many ways. I especially valued the hands-on work with computing systems and workflow design, which are skills highly relevant to my future career goals. I’m grateful for the opportunity to learn from and collaborate with such a supportive and knowledgeable team!
This summer, I worked at NSF NCAR in the Climate and Global Dynamics Lab. I utilized a large language model to help in converting antiquated code into modern languages and architectures. This code deals with converting ice sheet data during the process of running NCAR’s atmospheric models. In the end, I was able to create a draft of the new, updated script and ran tests to evaluate its functionality. Through this experience, I learned a great deal about file management and version control, using tools like Linux and GitHub. I was also able to explore the various possibilities of AI in software development and inspire new possibilities at NCAR!
I am beyond grateful for the opportunity to be at NSF NCAR for my summer internship. I was able to work with and learn from some extremely talented scientists and software engineers. Specifically I focused on improving part of the CESM (Community Earth System Model) post-processing workflow. This involved using NCAR’s high-performance computing systems, Derecho and Casper, to run simulations and process large climate datasets. My work centered on enhancing the CUPiD (CESM Unified Postprocessing and Diagnostics) framework by integrating diagnostics from both the Atmosphere Diagnostics Framework (ADF) and the International Land Model Benchmarking (ILAMB) package into a single, unified summary page. By consolidating outputs from these two tools, the project aimed to make it significantly easier for scientists to access, interpret, and compare complex model evaluation results.
During this summer, I had the privilege of working at NCAR under the guidance of expert scientists at the Mesa Lab. I was introduced to new topics within the Community Earth System Model (CESM) and CESM Unified Postprocessing and Diagnostics (CUPiD) via JupyterHub on their high-performance computing (HPC) cluster, Casper. Through this experience, I deepened my understanding of parallel computing and acquired new skills in Python libraries NumPy, Pandas, Xarray, and Dask beyond what I had encountered in my university coursework. The first half of my internship focused on conducting in-depth research and experimenting with difference plots to compare fields of data and apply these computations to extreme weather tracking software. Based on the uniqueness of the best approach for memory optimization with the intended softwares, the second half was more focused on working with Dask separately to understand the intricacies of dask parallelism for large-scale data analysis.
Summer 2024
Ingrid Carlson - NSF NCAR
I had an educational and enjoyable summer working with NCAR. The summer started with me learning about their Community Earth System Model (CESM) and CESM Unified Postprocessing and Diagnostics (CUPiD). The goal of my project was for CUPiD to run automatically with the CESM workflow. After several weeks of learning new skills and working with scientists at NCAR, I was able to successfully complete my project and my changes to the CESM workflow will hopefully be added in the near future.
Matthew Chung - Globus Labs
During the summer, I had the opportunity to work at Globus Labs where I got a chance to take a deep dive into Parsl, a Python based parallel scripting library used to parallelize workloads on supercomputers. My project was focused on researching and developing workload distribution algorithms in order to maximize the efficiency of Parsl’s High Throughput Executor. During this project, I learned how to develop, test and deploy the various algorithms using realistic workloads.
Cailin Cobey - OOI
This summer, I had the opportunity to work with WHOI on a project where I utilized HTML, CSS, and Beaver Builder within WordPress to create a professional dashboard that integrated data from Google Sheets and Google Drive. My project involved developing a centralized resource where users could access a calendar and view important datasets. By writing custom HTML and CSS, I was able to design a clean, responsive layout that enhanced the user experience. I used the Google Sheets API to pull real-time data into the dashboard, which was then displayed in tables and charts for easy analysis. The calendar was embedded using an iframe, ensuring seamless integration with Google Calendar. Overall, this project allowed me to blend front-end development skills with data integration, resulting in a polished and functional tool for users. I really enjoyed my time working with WHOI’s team and am very grateful for the opportunity!
Ellie Fahey - NSO (CICF 2023)
This summer, I was an intern for the development calibration team at the National Solar Observatory (NSO) data center in Boulder, Colorado. The data center team works alongside scientists on the Daniel K. Inouye Solar Telescope (DKIST), to create and maintain the data pipelines necessary to make solar observations available to scientists worldwide. Within DKIST, there are four (eventually five) instruments which all process different aspects of solar information, ranging from near-infrared to spectrography and visual imaging. Each instrument requires its own unique set of calibrations, visualization tools, and data pipeline in order to result in workable data for scientific use. It is the job of the development calibration team to write the code that will calibrate for mechanical effects on each instrument's data. My work focused on the Diffraction Limited Near-Infrared Spectropolarimeter, or DL-NIRSP, and writing code that does detector image correction to account for stray light on the instrument’s detector.
Raymond James Gallant - NSF NCAR
This summer, I had the incredible opportunity to work with scientists and software developers at the NSF National Center for Atmospheric Research in the Climate and Global Dynamics Lab. I gained a broad learning experience, exploring topics such as software development, parallel computing, and containerization, along with other areas that were new to me. I worked with Python libraries like Pandas, Dask, and NumPy, which enriched my understanding. This experience has significantly impacted my career goals, sparking new interests in areas that were previously unfamiliar and not covered in my regular university studies.
Mark Onders - NSF NCAR (CICF 2023)
I had an amazing internship this summer at NSF NCAR! I was so fortunate to be placed right on CU Boulder’s campus and able to work in person at the MESA facility. Not only is the MESA lab itself beautiful, but I really enjoyed getting to meet lots of people who work for NCAR and experience, first hand, the collaborative work atmosphere there. I was even able to go the the Research Aviation Facility and see NSF NCAR’s C130 airplane! This in itself was an incredible experience, but it was even more interesting because the research I did this summer directly relates to this plane!
This summer I worked within the Application Scalability and Performance (ASAP) team, which is located within the Computational and Information and Systems Lab. My team is currently working on code that will be used alongside NSF NCAR’s new Airborne Phased Array Radar (APAR) system. APAR will be placed on the outside of the C130 plane to allow it to gather observational meteorological data while in flight, which is pretty awesome! The code that my team is working on is called SAMURAI, or more formally known as Spline Analysis at Mesoscale Utilizing Radar and Aircraft Instrumentation. During my summer, I was able to learn valuable skills which allowed me to submit jobs from this code on NCAR’s supercomputers, Derecho and Casper. I was also able to learn how to benchmark, or essentially time, this code’s performance and create several PRs on github to help progress this project. I also researched other tools and frameworks that were being considered added to SAMURAI.
Overall, this was a great experience, and I was able to learn valuable skills that I am grateful to have in my career going forward. I am grateful for CI Compass, NSF NCAR, and the National Science Foundation for making this opportunity possible!
Palina Pauliuchenka - OOI
Over this summer, I worked on two projects with OOI. For the first project, I learned a lot about data from OOI, NEON, ONC, AOOS, and other programs. I pulled data for specific instruments and parameters, such as CDT and sea water temperature, using their APIs and integrated all the data in a Jupyter Notebook. I correlated the data by determining sampling rates and coordinating timestamps, and documented the notebooks thoroughly so others could use them.
For my second project, I analyzed trends over time and correlated them with El Niño and La Niña events. This project allowed me to apply my knowledge in research and data analysis, and I learned a lot about El Niño, La Niña, and other environmental events in the process.
Quan Quy - MagLab
This summer at MagLab (the National High Magnetic Field Laboratory) has been an incredible learning experience for me. I developed a predictive model to forecast power supply failures for magnets at the MagLab facility, focusing on key features that could affect the magnets' health and failures. I also delved into time series forecasting to predict coil resistance deviation, which indicates the health of the magnets' coils, experimenting with various statistical and machine learning techniques to enhance model accuracy. It was eye-opening to see the real-world impact of this work on helping the facility with planning and maintenance scheduling. Additionally, I improved an audio classification program intended to detect hazardous conditions in magnet operations through audio frequency analysis and computer vision, making it more efficient and functional.
Mahee Shah - NSF NCAR(CICF 2023)
During my second summer at NCAR, I had the opportunity to work on the Community Earth System Model (CESM) with the Land and Ice team. My focus was on gaining proficiency with NCAR's newest HPC system, Derecho. This involved learning the intricacies of data management and job scheduling, crucial for running large-scale climate simulations.
In the latter half of my summer, I concentrated on running a simulation using the land and ice topography updater, a tool critical for tracking ice sheets in Greenland and Iceland. This involved configuring and executing a T-compset simulation forced by a B-compset output, focusing on the period from 2015 to 2100. The primary goal was to analyze the impact of atmospheric forcing on the Greenland Ice Sheet under a specific scenario. Through this process, I not only honed my skills in setting up and running CESM cases but also gained valuable insights into the dynamics of ice sheet changes and their implications for future climate scenarios.
Lisa Schulz - UNDERC
This summer I worked with the University of Notre Dame Center for Research Computing (CRC) and the University of Notre Dame Environmental Research Center (UNDERC). The overarching goal for my project was to create a lake water budget model that could be applied to the lakes of the UNDERC property to see how the lake volumes change over time. I had to learn how to create a digital terrain model from LiDAR data, and how to put together a mathematical model using equations from other scientist's previous efforts on related topics that I fit to match the needs and constraints of my project. I also learned how to use ArcGIS and I significantly improved my R programming skills, since R was the software I used to gather and interpolate data we needed for the model, and then construct the functions we need to run the program. The UNDERC property is in an area with one of the highest densities of lakes in the world, so a hydrological model like the one I am creating gives insight to water and nutrient flows which ultimately dictate the health of this critical ecosystem.
Connor Vessely - OOI
I primarily worked on computer vision projects during my internship at OOI. My first project was automating video summarization for over 46,000 videos captured by OOI’s video camera. In this project, I developed a process to select 9 frames from each video to represent the breadth of the video’s contents. I also developed a pipeline to identify blank images and work towards cleaning up the images in the OOI database. Most recently, I have been working on utilizing the newly operational GPUs at OOI.
Summer 2023 with NCAR
Raja Ali (CICF 2023)
This summer at NCAR has been nothing but transformative for me! I've been involved in enhancing the Ionosphere Dynamo Model, a sizable Fortran codebase set to be incorporated into WACCM-X, an advanced Earth atmosphere model. It's been an eye-opening journey learning the ins and outs of working in a research setting at a place as significant as NCAR.
From setting up a CI/CD Pipeline from scratch for the model for automated testing purposes to creating a namelist file that boosted the model's computational efficiency by 117% by eliminating recompilation time, I've gotten a real deep dive into the technologies used in Climate and Atmospheric Science, as well as what common technologies are being used in the cyberinfrastructure space by most major facilities. It's been a pleasant surprise to experience the sense of collaboration at NCAR. Everyone's eager to pool their knowledge and lend a hand, which creates a supportive environment despite the technically demanding nature of our work.
Edward Lin (CICF 2022)
At NCAR I was able to immerse myself in a research environment through meetings and activities. Interacting with scientists and engineers provided me with insight into what a potential career in research could look like. On top of that, I worked on a project involving the classification of weather tweets which combined my interests in natural language processing and the research done at NCAR.
Eddie Mayor (CICF 2023)
I have been observing and participating in the NCAR_NEON project, which connects powerful models with NEON's measurement network, bringing together ecology and earth science. The goal is to predict how Earth's systems will impact ecosystems. I’ve learned how NCAR gathers NEON data sets, processes them for model input, and uses data analysis and visualization techniques so that various audiences can understand their research.
Bhagya Ram (CICF 2023)
I'm working at the Atmospheric Chemistry Observations and Modeling division at NCAR. At my internship, I'm working on MusicBox, a web-based model that simulates the chemistry and transport of trace gases and aerosols in the atmosphere. I'm working on the model's front-end, utilizing Javasript, Django and React. I'm very thankful for this opportunity. I'm learning a lot, and I'm excited to draw on my experience in the future!
Mahee Shah (CICF 2023)
During the summer, I had the privilege of working at NCAR to explore the field of machine learning. The first part of my summer was dedicated to in-depth research and experimentation with various modeling softwares. The second half was spent deconstructing a project on Natural Disaster Tweets Classification, with the goal of comprehensively understanding its intricacies while attempting to replicate its outcomes.
Summer 2023 with OOI
Gareth Oram (CICF 2023)
I worked for WHOI designing a “scraping tool” to extract desired information from data logs. The data logs were transmitted through wires and wirelessly from buoys that collect oceanographic data. My program also functioned as a sorting tool to sort the desired information into four different categories.
Calloway Sutton (CICF 2023)
I worked on two main projects while at WHOI, upgrading the document management system and working on a way to search through OOI’s large video dataset. For the first project I got to learn about the various different DMS’ on the market such as Alfresco, Mayan, and SmartyPants as well as some potential alternatives to a DMS like Gitlab, Onbase and the large list of cloud drive providers. For the second project, I had the opportunity to use my previous knowledge in machine learning to work on making a system which could efficiently search through petabytes worth of videos with little compute.
This project is supported by the U.S. National Science Foundation Office of Advanced Cyberinfrastructure in the Directorate for Computer Information Science under Grant #2127548.