The Science Behind CIBO

Jenette Ashtekar is the VP of Product Management at CIBO, a science-driven software startup. Prior to joining CIBO, Jenette spent ten years as an academic soil scientist and environmental modeler developing algorithms to help predict soil and environmental properties across scales. While pursuing research at Purdue University, Jenette co-developed and commercialized a soil mapping technology used to better inform farmers’ management decisions. Her functional soil mapping technology is used today in a commercially available, farm management software application. Jenette is passionate about the development and promotion of university technology into practical, real-world software solutions and believes that today’s agricultural challenges can only be solved through the integration of many diverse sciences, technologies, and industries.

Bruno is a co-founder of CIBO Technologies. An internationally recognized agricultural systems scientist, he is a professor in the Department of Earth and Environmental Sciences at Michigan State University. His research concentrates on the understanding of spatial and temporal variability of crop yield, water, and nutrients.

What led you to pursue a career in data science? 

“Early on in my career as a software developer, I got engaged in data analytics and found the experience rewarding. There is an unexpected joy when working with data, the possibility of finding answers to interesting questions and hypotheses which is very fulfilling. 

Driven by this interest, I went on to complete my Master’s in data science at Indiana University, Bloomington. Post-graduation, I accepted a full-time job offer at CIBO as a Data Scientist.”

What is your favorite thing about being a data scientist? 

“The prospect of seeking and finding the truth. Statistics and data science have long driven scientific research leading to the discovery of new facts about the world. Being a data scientist gives me the privilege to contribute to the scientific advancements by leveraging data science skills and subject knowledge to model real-world systems and processes. These models can then be used to simulate different scenarios and help answer complex questions.”

What is one thing that would surprise people about your work in data science? 

“Although data science has come into vogue lately and has received increasing attention, I think there is a less formal acknowledgment of its roots in rational and native human thought processes.”

“The human mind is a great statistical machine. Our mind continuously and incrementally aggregates and summarizes observed inputs, as storable and retrievable knowledge blocks, that makes reasoning, and inferencing so seamless. Data science is analogous to the way our brains work and provides a formalized systematic framework to think about and apply. The Bayesian learning frameworks in Data science reflects these ideas very closely.”

What kinds of things do you think we should be doing to encourage more people to pursue a career in science? 

“We should provide opportunities for high school and middle school students to engage with scientific research. This can help them appreciate the process of trial and error and balance all outcomes. Academic institutions and industry can further augment this effort.”

What brought you to CiBO? 

“I graduated from Indiana University, Bloomington in 2017 earning a Masters degree in data science. CiBO appealed to me for its groundbreaking work combining science and Technology.”  

“Furthermore, I found the idea of applying technology in the agriculture sector enticing as farming is one of the oldest human professions yet most underserved by technology.”

What are your primary responsibilities at CiBO? 

“As a Data Scientist in the modeling team at CIBO, over the past two+ years I’ve worked extensively on our core crop models in validating and improving them. This has been done in collaboration with our expert cross-sectional team of crop modelers and scientists. I have worked on using statistical techniques to identify sources of errors and help identify potential model improvements and development. I am also responsible for programming these models into production deployable workflows.”

“More recently, I’ve been working on the Farmland valuation project that is chartered with goals to estimate the values of Farmland across the continental US. We employ state-of-the-art data science methods combined with our core simulation models to estimate farmland values.”

What do you find the most rewarding about your work here? 

“The diversity of skills at CiBO amazes me and inspires me to do my best and continuously strive to up the bar. We have experts in a range of diverse domains such as software engineering, crop science, soil science, computer vision, and data science. The learning potential cutting across these different domains is immense and enriching. It is very gratifying to know that we are tackling some of the biggest challenges in human history, and contributing to solve problems in the agriculture domain and help our farmers grow more sustainably and also help improve their livelihood scenarios by increasing profitability and maximizing outcomes.”

“I think CIBO is uniquely placed in terms of skills and capabilities needed to combine science and engineering at a large scale to develop a platform that can address some of the complex problems of our lifetimes.”

About Subhash Bylaiah

Subhash Bylaiah is a Data Scientist at CIBO, a science-driven software startup. He holds a Bachelors of Engineering from the Malnad College of Engineering and a Master of Science in Data Science from Indiana University Bloomington.

What brought you to CIBO? 

“I find the application of computer vision to satellite imagery and remote sensing at CIBO exciting. In particular, using computer vision and machine learning techniques to understand crop fields is extremely interesting. Additionally, CIBO was looking to use wide range of different approaches to solving these kinds of problems, from Bayesian statistical models to deep learning, which lined up perfectly with my Ph.D. and academic research experience.”

What inspired you to pursue a career in data? 

“My academic training is in traditional computer vision. When I first started my work in this field 15 years ago, it was a relatively new area with a huge number of problems yet to be solved. Although we have made incredible progress since then, there are still many challenging computer vision problems that need solutions that can be applied in the real world. In particular, computer vision applied to remote sensing is very new and we have only scratched the surface of what can be accomplished.”

What is your favorite thing about being a computer scientist?

“The best part of being a computer scientist in computer vision is working on incredibly challenging problems that have the potential to impact the future in a very real way. In particular, working at CIBO has given me the opportunity to tackle these problems in a highly collaborative, multidisciplinary environment, which makes this type of work even more enjoyable.”

What is one thing that would surprise people about your part of computer science? 

“The fact that computer vision, as a scientific problem, is incredibly difficult. I think it is hard to grasp this because human vision is extremely “easy” for us; we make sense of the world through what we see without seemingly any effort. However, the underlying processes controlling vision are extremely complex and took millions of years of evolution to get right. “Solving” computer vision would mean essentially replicating this ability in computers — a very challenging task!”

What are your primary responsibilities at CIBO? 

“My main role is to design and develop algorithms for understanding remotely sensed imagery, e.g., what can we say about a crop field given some satellite images of it. Part of my responsibility is also to understand the impact of solving each problem on business outcomes for CIBO, as well as on the work of other scientists here, with the ultimate goal of moving the remote sensing work in a direction that is beneficial to the company.”

What do you find the most rewarding about your work here? 

“The problems we solve at CIBO are extremely challenging and exciting, with lots of potential societal impact. Also, the people here are extremely talented and so much fun to work with.”

Published articles 

For further articles by Ernesto Brau, visit his Google Scholar Page: https://scholar.google.com/citations?user=e8qul0UAAAAJ&hl=en 

About Ernesto Brau

Ernesto Brau is the Lead Computer Vision Scientist at CIBO, a science-driven software startup. Prior to CIBO, he worked on computer vision for AiBee Inc., Intel Corporation, and Hewlett-Packard Laboratories. He holds a Bachelor of Science in Computer Science from the Universidad de Sonora in Mexico, along with an MS and PhD in Computer Science from the University of Arizona.

What led you to astronomy and how does it tie into your job at CIBO? 

“In high school, I took a telescope out to my backyard and was hooked by looking at the stars. It seemed amazing to me that people were actually paid to stargaze. I decided I wanted to be one of those people!”

“My work in astronomy was always more observational in nature rather than theoretical. I would go to a telescope, observe an astronomical object, analyze the resulting data and publish my findings. What I enjoyed most was talking to other researchers about how my conclusions fit in (or did not fit in!) with their findings. I also enjoyed writing code to clean and analyze data. I engage in very similar activities at CIBO, spending the majority of my time cleaning and analyzing data, as well as sharing my conclusions with others on the team.” 

What is your favorite thing about being a data scientist? 

“The initial discovery phase is my favorite because you don’t know exactly what you’re going to find. It usually occurs after you have done the cleanup work that’s critical but not always the most exciting. This is where your key contribution will be — to figure out what’s important in the data. You find key insights that weren’t known before and you get to share them with others. What could be better than that!”

What is one thing that would surprise people about data science? 

“The sheer amount of that effort must be expended on pre-processing and data cleaning. It almost always ends up being a majority of the time and effort when you tackle a new data science project.”

What brought you to CIBO? 

“I was drawn to CIBO because I wanted to have a positive impact with my work. CIBO has a strong focus on changing the way we understand agriculture in a way that will be beneficial across many sectors of the agricultural economy. In addition, from a technical perspective, it was also appealing to me to learn Scala and to be expected to use Bayesian analysis techniques in my work.”

What are your primary responsibilities at CIBO? 

“Right now, my task is to get our cotton model working smoothly across the country. I’m leading a small team and making sure that the work we are doing is aligned with the goals and needs of CIBO’s core product offering. I also make individual contributions to the team, primarily handling our data ingestion and model validation tasks.”

“In the past, I’ve worked on the remote sensing team and helped to make contributions to our proprietary CIBO reliability score.”

What do you find the most rewarding about your work here? 

“It’s a combination of being intellectually stimulated and having a positive impact on the environment. Especially with the sustainability emphasis, I believe that we’re tackling challenging but tractable problems. My work has a positive impact on the world!”

About Shane Bussmann

Shane Bussmann is a Senior Data Scientist at CIBO, a science-driven software startup. Prior to CIBO, he worked as the Lead Data Scientist for Understory, Inc. He holds a B.A. in Physics and Astrophysics, along with a Ph.D. in Astronomy from the University of Arizona.

What led you to data science? 

“My interest has been in applying data science skills to different problems that primarily fall in a scientific domain. I wanted to take data and apply it to scientific applications. Before CIBO, I was pursuing a Masters at the University of Massachusetts Amherst, and as part of my research, I worked with a team of biologists at Cornell to help them study how birds migrate across the US using weather radar data. We were trying to identify the roosting locations of tree swallows to understand their migration patterns throughout the year. The goal was to build tools that allow us to monitor bird populations and how various environmental factors might impact them.”

“I also like learning about astronomy and have done project work with data obtained from modern telescopes to help identify astronomical objects such as stars, galaxies, and supernovae. Astronomical datasets are often terabytes in size and you need machine learning-based tools to process these datasets and generate meaningful results. I also volunteer with a research lab in astronomy where we use machine learning to study gravitational lensing from telescope images. This in turns helps improve my data science and computer vision skills.” 

What is your favorite thing about being a data scientist? 

“I transitioned into this field less than a year ago. I initially joined CIBO as a software engineer, then transitioned into a data scientist. I’m new to the field, so the process itself is exciting – coming up with a question to answer, looking at how we can use the data to test our hypothesis, performing analysis to get results, and finally creating visualizations to communicate those results.”

“For example, here in the remote sensing team, we are working to improve our reliability maps. When we make some improvements to our model we try and generate visualizations, such as a map of reliability scores for multiple states in the US. We use these maps to explain the high-level insights. For instance, the presence of a water body, such as a river nearby may improve the reliability of that land and result in a higher score. The visualizations make it easier for people to understand these ideas that may otherwise be difficult to communicate.”

What is one thing that would surprise people about your field of work in data? 

“As someone without a background in agriculture, I’m surprised how data-intensive the field is. I was amazed to find that combines are equipped with sensors to track its location on the field as well as to measure the rate at which grains are falling into the combine. Family farm groups that hire expert agronomists, whom I call farmer data scientists, to work with the data, clean it and process it. So clearly farming is a lot more complicated than planting and harvesting, which is what my naive self thought.”

What kinds of things do you think we should be doing to encourage more people to pursue a career in science? 

“I think it would definitely help to spend more effort on public outreach and education about the importance of science in our daily lives. I believe there’s never been a better time to be a scientist than now. We are connected as a species like never before, thanks to the internet. We have the resources to solve important problems in the world, thanks to technology. We just need to be creative in our minds and be motivated to produce scientific solutions to those problems. This is not always easy because of the excess of bad/negative things shown in today’s media. We as individuals need to realize that there are a lot of good things happening, especially in science. We are making new breakthroughs, inventing new cures for diseases, designing new technologies that can fundamentally change our lives. Encouraging people to adopt a positive mindset towards science would be a good way to have them like the field and want to get involved.”

How are you involved in the local science community? 

“I am one of the organizers of a local data science meetup group, very creatively named the Boston Data Science meetup. Once a month, I lead a book club meeting, where we pick a book on machine learning that we are interested in reading, go through it and discuss our thoughts about it. The book club is reasonably popular, at one point, we had over 40 people joining us. It helps to build a community of people from different industries and network around data science topics.”

What brought you to CIBO? 

“I came to CIBO because I was interested in applying data to scientific applications. The focus of the company in agriculture aligned well with my career aspirations. I also came because I had just graduated and was looking for a job, but I stayed because the work environment was conducive to research and growth and it aligned with my personal goals of working in a scientific field.”

What are your primary responsibilities at CIBO? 

“I’m a Data Scientist on the Remote Sensing team. I work on improving our remote-sensing products. I’ve chiefly worked on stability/reliability maps and expanding their reach to include crops like cotton and wheat. I’ve also worked on tools that allow us to generate these maps and scores at scale across the US. Other than that I’m also involved in the Data Science Journal club, where we look at how we apply the latest research in Data Science, Machine Learning and Computer Vision to build tools that will help CIBO in the long term.”

What do you find the most rewarding about your work here? 

“The freedom to pursue independent research – bring your own ideas on how you can improve the product and get a chance to explore them for a reasonable length of time. CIBO also provides the opportunity for personal development and shape your career the way you want it.  I am from an Engineering background, transitioning into Data Science. I really wanted to improve my stats skills since it was important for my current job and future career. CIBO has allowed me to dedicate time to pursue online courses and even sponsored course subscriptions wherever needed.”

About Pankaj Bhambhani 

Pankaj Bhambhani is a Data Scientist at CIBO, a science-driven software startup. He holds a B.Tech in Information and Communication Technology from Dhirubhai Ambani Institute of Information and Communication Technology and a Masters in Computer Science from the University of Massachusetts Amherst.

What led you to crop modeling? 

“As a student majoring in Environmental Engineering, I was introduced to mathematical modeling during my Master’s degree in India. I learned my mathematical modeling instructor had worked with Dr. Stephen Hawking at Cambridge, so I tried very hard to learn from him beyond the classroom lectures. It was my instructor, Dr. Sat Ghosh, who ignited my curiosity in systems analysis and modeling.” 

“My first independent modeling project was on why the marble on the Taj Mahal was becoming yellow. It was challenging, but the results from the acid rain modeling were very meaningful. Another important milestone in my student life was working on the Gaussian plume modeling on air pollution. Our research results helped engineers make decisions on where safe areas were constructing factories and how to consider the extent of dispersion of pollutants in the environment. I also worked on biogeochemical models for my Ph.D. It was during my stint as a postdoctoral researcher that I began to work with crop models.”

“You know that feeling when you love what you do? I felt like that every time I took up a process model project. The surprise element in modeling is what keeps me going.”

What is your favorite thing about being a scientist? 

“I get to work with real world problems. Our research helps in developing decision support tools for farmers. I work with very smart data scientists, statisticians, agricultural scientists and engineers. It gives us a platform to think together, work together and solve problems. The diversity in our background and specialization is my favorite part of working as a scientist.” 

What is one thing that would surprise people about your field of work in agriculture and data? 

“We work with many different complex mathematical, statistical and systems analysis methods. The crop models need a fair understanding of linear algebra. I don’t think people always relate agriculture to complex mathematics. The buzzword “AgTech” involves many computer programming, remote sensing, mathematics and statistics involved than most people presume it to be. As a process modeler, I integrate mathematics into cropping systems and soil processes and help in better decision making. Predictive modeling in agriculture is challenging yet very interesting.”

What kinds of things do you think we should be doing to encourage more people to pursue a career in science? 

“We should introduce computer coding right from middle school. You may disagree with me by saying that coding might not be of interest to every student but it opens up a plethora of opportunities for students. Today, multiple disciplines need coding for better probing into their research areas. A wildlife ecologist, an agronomist or a hydrologist all need to have a fair grip on coding. Academic institutions and industry can partner together and help in curriculum design where kids get hands-on coding experiences.” 

Have you won any awards? 

“I have won several performance-based scholarships throughout my Ph.D. including The Grinter Fellowship, awarded to outstanding international students and the Office of Research fellowship at the University of Florida. I’m also a member in the Gamma Sigma Delta honor society of agriculture for my academic performance.” 

Selected Papers 

Chatterjee, Nilovna, et al. “Simulating Winter Rye Cover Crop Production under Alternative Management in a Corn‐Soybean Rotation.” Agronomy Journal, 26 July 2020, doi:10.1002/agj2.20377. 

Chatterjee, Nilovna, et al. “Changes in Soil Carbon Stocks across the Forest-Agroforest-Agriculture/Pasture Continuum in Various Agroecological Regions: A Meta-Analysis.” Agriculture, Ecosystems and Environment, vol. 266, Nov. 2018, pp. 55–67. 

Chatterjee, Nilovna, et al. “Biochar in the Agroecosystem–Climate-Change–Sustainability Nexus.” Frontiers in Plant Science, vol. 8, Dec. 2017, doi:https://www.frontiersin.org/articles/10.3389/fpls.2017.02051/full. 

Chatterjee, Nilovna, et al. “Depth-Wise Distribution of Soil-Carbon Stock in Aggregate-Sized Fractions under Shaded-Perennial Agroforestry Systems in the Western Ghats of Karnataka, India.” Agroforestry Systems, vol. 94, no. 2, 2019, pp. 341–358., doi:10.1007/s10457-019-00399-z. 

Chatterjee, Nilovna, et al. “Decadal Changes in Shoreline Patterns in Sundarbans, India.” Journal of Coastal Sciences, 15 July 2015, drs.nio.org/drs/bitstream/handle/2264/4847/J_Coast_Sci_2_54.pdf?sequence=1. 

View Nilovna Chatterjee’s full Google Scholar profile: https://scholar.google.com.br/citations?user=9jh6uIEAAAAJ&hl=en

What brought you to CIBO? 

“CIBO’s technology. Very few companies amalgamate technology, agricultural science and engineering at a large scale as smoothly as CIBO. As a crop modeler, I was aware of CIBO’s modeling platform and how they scaled it up to a whole new level. I joined CIBO because their work in the modeling teams aligned with my interest. It makes work easier when you love what you do.”

What are your primary responsibilities at CIBO? 

“I am a crop modeling scientist with the Data Science and Modeling team. My primary responsibilities include developing new crop models, improving model performance, and testing and validating crop model performances. A significant part of my work involves integrating data science techniques to improve the capabilities of our models.” 

What do you find the most rewarding about your work here? 

“I get to work on issues in agriculture that I deeply care about at CIBO. Engineers and scientists work together to solve a common problem. My coworkers are very friendly and helpful. It makes a huge difference when you work in a healthy and happening environment such as CIBO’s.”

About Nilovna Chatterjee 

Nilovna Chatterjee is a Crop Modeling Scientist at CIBO Technologies, a science-driven software startup. Prior to CIBO, she worked for the University of Nebraska-Lincoln, department of Agronomy and Horticulture as a postdoctoral research associate. She holds a Bachelor of Engineering, Master of Engineering from Vellore Institute of Technology, Vellore, India and a Ph.D. in Forest Resources Conservation and Soil Science from the University of Florida, Gainesville.

Marie Coffin is the VP of Science & Modeling at CIBO, a science-driven software startup. She has focused on being a biostatistician at agriculture companies. Prior to CIBO, she worked for Monsanto, Icoria, Paradigm Genetics, and was an assistant professor at Clemson University. She holds a BS in Mathematics from South Dakota State University and a Ph.D. in Statistics from Iowa State University.

What brought you to CIBO? 

“It was CIBO’s technology. CIBO’s crop modeling platform includes models whose development started at Michigan State University. The company is truly a knowledge hub, continuously building upon scientific research it conducts over time to address real-world problems. I knew CIBO’s platform was a powerful piece of science and engineering, and there is a wide range of applications that can be based on it. I joined CIBO because I believe there are lots of things I can offer to make this technology even better.”

What is one thing that would surprise people about your field of work in agriculture/data? 

“One thing that surprises people is the number of advanced mathematical, statistical, and systems analysis methods that have been adapted to agricultural sciences. Agriculture has not been traditionally a maths-oriented field. However, methods developed in other disciplines (e.g., biology or engineering) have been adopted by a certain group of scientists to make agricultural decision-making more efficient. When most folks think of agriculture, they imagine soil science or crop science, but actually, there is much more to the field. Mathematical modeling, for example, is widely used in agricultural sciences to support decision making.”

How did you get into your line of work? 

“It was during my last year at the School of Agronomy in Togo, West Africa that I became interested in crop modeling. I attended a seminar on systems dynamics during which the speaker demonstrated a different perspective on analyzing crop growth using simulations—that was very different from what is usually done in traditional agronomy. The speaker explained how systems analysis methods developed in industrial engineering could be applied to agricultural systems to enhance the decision-making process. By considering crop growth, for example, as occurring in a system and placing the problem to solve at the center of the system, one can develop multi-disciplinary solutions using simulation models as tools. I was fascinated by this approach and decided to continue my studies in systems analysis and crop modeling.”

What are your primary responsibilities at CIBO? 

“I contribute to the development of new crop models, along with testing and validation of existing crop models. My work involves expanding the capabilities of our crop models to include other data science techniques. I help make the crop model outputs more useful to our customers.”

What is your most memorable moment at work? 

“My favorite times at work are when I get together with other colleagues from different backgrounds to work on projects. We dig deep into solutions and try to find the best and practical ones. I love deep scientific curiosity. Once you start digging into details, you can find things you don’t expect. I love doing that. It keeps the scientific curiosity alive. Working with people with different backgrounds gives you the opportunity to hear different points of view and try methods you may not have thought of on your own.”

“You don’t always find this level of scientific curiosity and collegiality at other companies. CIBO encourages employees to deep dive into issues and find impactful solutions.”

What do you find most rewarding about your work here? 

“I get to work on the most pressing issues in agriculture of our time and collaborate with very clever people from diverse backgrounds and points of view. It’s rewarding because sometimes you see problems or solutions only from your perspective. Someone who brings a different perspective to an issue can help you learn from their expertise—and broaden your views.”

What kinds of things do you think we should be doing to encourage more people to pursue a career in science? 

“I think Universities should offer courses that prepare students for careers that aren’t necessarily academic. A significant proportion of students in the agricultural sciences may not want to stay in academia—they might be interested in a career in the private sector, for example. The jump from school with a degree in science to a non-academic environment is not trivial. Helping students bridge that gap very early on would go a long way.”

What advice would you give to young people interested in a career in science? 

“It’s a common perception that science is very abstract and theoretical. Students might think that pursuing a career in science outside of academia is hard—but this is not necessarily true.”

“I think it would be useful for young people interested in a career in science to do an internship with a company that they like. It’s important to read job descriptions to get a feel for the kind of positions available, the kind of skills companies are looking for, or even unresolved issues they might want to tackle through entrepreneurship—all these while still in school. They can then re-shape their academic programs to suit the current requirements of the job market better.”

What are your hobbies and do they influence your work? 

“In my spare time, I enjoy going for runs and pretending like I’m preparing for marathons. I also love reading popular science books because they help me learn techniques to distill science into easy-to-understand and interesting stories.”

What do you hope to see in your field in the next 10 years? 

“The digital ag sector is slowly emerging. While the ag sector is a bit resistant to new technologies, I hope to see a high adoption rate of these technologies in the near future. I want the agriculture sector to accelerate its transition to the digital age and catch up with other industries.”

Why are you excited about CIBO? 

“In terms of the digital ag sector that CIBO is part of, I believe one hundred percent in the company. I think CIBO has the best technologies, the best platform, and the right strategy. This combination really gets me excited!”

 

About Kofi Dzotsi

Kofi Dzotsi is a Principal Crop Scientist at CIBO Technologies, a science-driven software startup. Prior to CIBO, he worked for The Climate Corporation as a crop simulation modeler and quantitative researcher. He holds a degree in agricultural engineering and agronomy from the University of Lomé along with an MS and Ph.D. in agricultural and biological engineering from the University of Florida.

What led you to software engineering? 

“When I was in elementary school our school had a Radio Shack TRS-80 that came with a built in BASIC interpreter. I spent most of my free time figuring out how to write programs for it. Ever since then I’ve been fascinated with computers and writing software.”

What is your favorite thing about being a software engineer

“One of my favorite things about being a software engineer is the flexibility to apply those skills to a lot of different domains. Over the course of my career, I’ve worked in a number of different industries. I’ve always been broadly curious so it is fun to be able to work on many different problems while still developing a specialized technical skill set.”

What is one thing that would surprise people about software engineering?  

“One thing that surprises people who are not directly involved in software engineering is how much of it there is everywhere. For example, people are often surprised by how much software/technology there is in farming.”

What brought you to CIBO? 

“The opportunity to work at a company that is basing its product on actual science in a fundamental way. Agriculture is interesting because it’s an important, productive part of the economy and there are opportunities to build something more meaningful than the next social media app. I also knew there were really smart teammates here who would be great to work with.”

What are your primary responsibilities at CIBO? 

“I am leveraging my software engineering expertise to help the science and modeling team deliver data-driven insights at scale.”

What do you find the most rewarding about your work here? 

“One of the most rewarding things has been working with experts from a variety of fields. I’m impressed and humbled by the depth of knowledge everyone brings to the team. Watching as we come together to generate new information and insights about farmland around the US is pretty amazing.”

About Jeremy Keillor

Jeremy Keillor is a Senior Software Engineer at CIBO, a science-driven software startup. Prior to CIBO, he worked for Virgin Pulse, RedBrick Health, and Godengo. He holds a Bachelors degree in English Literature and Computer Science from Bethel University.

What brought you to CIBO? 

“I was looking for a small company because I wanted to work at a faster pace. I really liked CIBO’s focus on sustainability. As a society, we have the data and knowledge to better manage our land. I wanted to work for a company that had a mission that resonated with me. CIBO fits into my skills really well with the combination of biology, agriculture, and data science.”

What led you to data science?

“Ecology led me to data science. I’ve always loved the natural world but didn’t know what jobs were in this space. As a girl who was interested in science and math, I was steered toward computer science. My parents were also computer scientists and because I was inclined towards technology, this was a good fit initially.”

“While working in the computer world in Washington, DC, I became interested in environmental justice and land use. I learned about alternative business models for agriculture, including Community Supported Agriculture. Eventually, I left my computer job and went to work on an organic vegetable farm. But I missed research, so I found a way to combine my environmental interests with my background in computer science. I decided to pursue an advanced degree in ecology with a focus on the effects of human-caused change to the environment.”

“Part of my Ph.D. focused on beef production in the US and the economics behind it. Other parts of my dissertation concentrated on plant ecology and biodiversity. My advisor’s background was in plant ecology, but he had recently become focused on sustainable food production and the question of how to feed 9 billion people in an environmentally sustainable way.”

“It turned out that my background in computer science was useful because a lot of ecologists don’t have a background in computer science. I brought this combination of skills to CIBO because I was attracted to the mix of complex problems, agriculture and computer science within my role.”

What are your primary responsibilities at CIBO? 

“I lead the sustainability team. My team’s job is to develop algorithms to estimate the environmental impact of farming on agricultural land throughout the country.”

What is your favorite thing about being a scientist? 

“I enjoy understanding and explaining some of the complexities of the biological world.” 

What is one thing that would surprise people about your field of work in agriculture/data? 

“The thing that surprised me the most was how much we don’t know about the natural world. We don’t know what all the functions of a cell are. We don’t know why zebras have stripes. We don’t know some very fundamental things that people assume scientists know. We’re better at understanding cars and computers and other built things than our own bodies and the environment we inhabit.”

What kinds of things do you think we should be doing to encourage more people to pursue a career in science? 

“The most important thing is to teach kids that science is a process of exploration and not a set of facts. We have textbooks and tell kids to memorize the content. I didn’t take a biology course until I was in graduate school because to me biology meant memorizing things about cells and molecules and organic chemistry. I didn’t want to memorize things I could look up.” 

“Science is not a list of facts and things we know. It’s about asking questions and trying to figure out the answers. That is fun for kids. I think when kids learn science in a well-taught process-based framework, it hooks into their natural curiosity and engages them.” 

“I also find that people perceive a bigger gulf between scientists and nonscientists than is actually there. For several years, I built and managed citizen science projects that brought anyone who wanted to volunteer into the science fold by having them participate in real science projects with scientists. I highly recommend anyone interested in participating directly in science research visit the Zooniverse, which hosts many such projects and with which I collaborated.” 

What do you find the most rewarding about your work here? 

“I find it rewarding to have my work incorporated into the product, the application that is seen and used by other people. In academia, I worked on the impacts of global change, and especially climate change. That’s important work, but you don’t necessarily see the results. You can write a paper and send it out into the world but you don’t know who’s reading it and if it will impact policy. But here, I come up with an algorithm, write the code, and can see how it impacts the product.”

Press Coverage:

• Undark – Taking the Pulse of the Planet
• Wired – Drones Are Turning Civilians Into an Air Force of Citizen Scientists
• Science Daily – Use artificial intelligence to identify, count, describe wild animals
• Venture Beat – Researchers develop AI that identifies and counts wildlife with 96.6% accuracy
• Wired – The Animals of the Serengeti Get Caught in Surprise Selfies
• Hakai Magazine – Citizen Science Comes of Age
• Buzzfeed – 23 Animal Selfies For Everyday Situations

Speaking Events 

• 2016 – Brown University, Providence, RI. Department of Computer Science. Data science, the environment, and the future of our natural world.
• 2016 – Arnold Arboretum, Boston, MA. Botany Blast: Observing nature for citizen science.
• 2015 – ComSciCon National Workshop on Science Communication, Cambridge, MA. Panelist for Multimedia Communication for Scientists.
• 2014 – Smithsonian Institution, National Museum of Natural History, Department of Entomology, Washington, D.C. Effects of experimental warming on a grassland insect and spider community.
• 2013 – Adler Planetarium, Chicago, IL. From lions to black holes: How citizen scientists are changing the face of research.

Selected Papers

Catford, J.A., A.L. Smith, P.D. Wragg, A.T. Clark, M. Kosmala, J. Cavender‐Bares, P.B. Reich, D. Tilman. (2019) Traits linked with species invasiveness and community invasibility vary with time, stage and indicator of invasion in a long‐term grassland experiment. Ecology Letters, 22(4). doi:10.1111/ele.13220

Kosmala, M., K. Hufkens, A.D. Richardson. (2018) Integrating camera imagery, crowdsourcing, and deep learning to improve high-frequency automated monitoring of snow at continental-to-global scales. PLoS ONE, 13(12): e0209649. doi:10.1371/journal.pone.0209649

Richardson, A.D., K. Hufkens, T. Milliman, D.M. Aubrecht, M. Chen, J.M. Gray, M.R. Johnston, T.F. Keenan, S.T. Klosterman, M. Kosmala, E.K. Melaas, M.A. Friedl, S. Frolking. (2018) Tracking vegetation phenology across diverse North American biomes using PhenoCam imagery. Scientific Data, 5:180028. doi:10.1038/sdata.2018.28

Kosmala, M., A. Wiggins, A. Swanson, B. Simmons. (2016) Assessing data quality in citizen science. Frontiers in Ecology and the Environment, 14(10):551–560. doi: 10.1002/fee.1436

Kosmala, M., A. Crall, R. Cheng, K. Hufkens, S. Henderson, A.D. Richardson. (2016) Season Spotter: Using citizen science to validate and scale plant phenology from near-surface remote sensing. Remote Sensing, 8(9): 726. doi:10.3390/rs8090726

Kosmala, M., P. Miller, S. Ferreira, P. Funston, D. Keet, C. Packer. (2016). Estimating wildlife disease dynamics in complex systems using approximate Bayesian computation models. Ecological Applications, 26(1): 295-308. doi:10.1890/14-1808

Swanson, A., M. Kosmala, C. Lintott, C. Packer. (2016) A generalized approach for producing, quantifying, and validating citizen science data from wildlife images. Conservation Biology, 30(3): 520-531. doi:10.1111/cobi.12695

Mosser, A., M. Kosmala, C. Packer. (2015). Landscape heterogeneity and behavioral traits drive the evolution of lion group territoriality. Behavioral Ecology, 26(4): 1051-1059. doi:10.1093/beheco/arv046

For further articles by Margaret Kosmala, visit her Google Scholar Page: 

https://scholar.google.com/citations?user=6dBv_6gAAAAJ

About Margaret Kosmala

Margaret Kosmala is a Senior Data Scientist at CIBO, a science-driven software startup. Prior to CIBO, she was a postdoctoral fellow at Harvard University and a predoctoral fellow at the Smithsonian Institution’s National Museum of Natural History. She holds a Ph.D. in Ecology from the University of Minnesota and a Bachelor of Science in Computer Science from Brown University.

What is one thing that would surprise people about your field of work in meteorology? 

“People understand that the weather is important, but many only view meteorologists as that person you see on TV. However, there are many other opportunities for meteorology jobs. Meteorology is the study of weather.”

“A part of meteorology people don’t often consider is that it’s not just the exciting weather (severe storms, tornados, hail, heavy rains, and hurricanes) that impact agriculture. The “nice days” (hot clear skies light winds) can be just as impactful to crops depending on the time of the year. Weather directly and indirectly impacts people’s lives.”

“Throughout my career, I’ve chosen roles that put me at the forefront of impactful weather both in public health (working in applied air quality and meteorology research) and feeding the world (working in ag-tech). Large organizations throughout the agriculture chain have positions that require an expert in meteorology to gather, process, and analyze weather data and its impact. There are also opportunities for research, forecasting, and consulting.”

How did you get into your line of work? 

“I knew I wanted to enter meteorology long before high school. Even in middle school, I enjoyed observing storms and learning about the weather. As I learned more about the field, I realized that meteorology blended my passions for math, physics, science, and computers. In college, I took many of the same core classes as the engineers and physicists. These classes came in handy as much of meteorology is solving and modeling the atmosphere using complex differential equations and linear algebra.”

What is your most memorable moment at work? 

“The most memorable moments have come from being able to work with people of different backgrounds to solve complex problems. From agronomists to software engineers, each person brings their unique perspective on how a problem could be solved and what the physical impact would be on land.”

“When examining how to model weather, you can’t just use software engineering and machine learning. Models need to take into account multiple academic and scientific disciplines to be credible and accurate. Working with cross-disciplinary teams allows us to create dynamic science-based solutions.”

What kinds of things do you think we should be doing to encourage more people to pursue a career in science? 

“Meteorology is a field where the job opportunities aren’t as commonly understood. It’s on the industry as a whole to help educate students in undergraduate and graduate programs that there are options outside of academia or the National Weather Service. There are plenty of opportunities in startups, private industry, or in the tech field that regularly open up, but they’re harder to find and most college programs are simply unaware of these opportunities.”

“I found my first job on my own, but it would have been helpful to have better job boards with these types of roles. This is why I look for any opportunity to mentor and educate students on other possible alternative job opportunities.”

What advice would you give to young people interested in a career in science? 

“Two key skills you should have: 

• A strong scientific programming background, such as Python; and excellent professional scientific communication skills;  and
• A strong ability to tell a compelling story with your science. Many scientists struggle to share their research in a compelling way to people who aren’t scientists. It’s essential to learn how to do effective professional scientific communication to present your science. If you can clearly communicate your ideas, you won’t have a problem finding a job.”

What do you hope to see in your field in the next ten years?

“A better general understanding of the global impacts of climate change and what we are doing to the planet. The science is “clear” and there is a consensus among scientists that Climate Change is real and is manmade. The wets are getting wetter, the hots are getting hotter, storms are getting stronger and more frequent.  My hope is that in ten years the populus will be listening to the science community and have started taking actions to forced actions aimed at reducing and mitigating the impacts of climate change globally. I hope there will be better partnerships between government, private, and academic communities to understand and address Climate Change’s impact on society.”

What brought you to CIBO? 

“I joined CIBO because of the importance placed on the weather. CIBO recognized the importance of having an expert in weather for our proprietary technology platforms. I was excited about the promise to be that weather expert.”

Why are you excited about CIBO? 

“The vision and capabilities of the technologies at CIBO are unprecedented. We have the opportunity to be a disruptive force when it comes to looking for and valuing land. The possibilities are endless, and it’s on us to capitalize on it. I’m excited by the importance CIBO has placed on solutions backed and supported by science.”

What are your primary responsibilities at CIBO? 

“I’m the weather guy, and I help to make sure the weather data is used in a scientifically correct way. I’m helping develop short and long term strategies for CIBO’s weather needs to improve CIBO’s products and services.”

What do you find the most rewarding about your work here? 

“I’ve enjoyed blending a role of theoretical science with software engineering. It’s refreshing to have colleagues recognize the importance of crafting a strategy for weather data; and being able to build a science-based software framework where weather data is being used correctly and in the best way possible. It’s rewarding to know that the work you do matters and is appreciated by the company.”

About Adam Pasch

Adam Pasch is the Principal Data Scientist of Meteorology at CIBO, a science-driven software startup. Dr. Pasch is a Certified Consulting Meteorologist by the American Meteorological Society as an expert in the application of weather information. He is an experienced Environmental Scientist with a demonstrated history of working in the scientific-software and applied research industries. Prior to CIBO, he worked for The Climate Corporation focusing on Weather Data Strategy and Operations and at Sonoma Technology, Inc as a Meteorologist and Project Manager. He holds a Ph.D., M.S., B.S. from Saint Louis University in Meteorology.

What brought you to CIBO? 

“What attracted me to CIBO was the extra attention that both the science and engineering teams paid to their craft. At CIBO there was a strong culture of principled approaches, including functional programming, Bayesian statistics, science-based modeling, and solid software testing. The teams were highly collaborative, blending engineers, scientists, and data scientists together, which allowed me to combine my interests in both coding and data analysis.”

What inspired you to pursue a career in data science? 

“I’ve always enjoyed solving challenging problems. After earning a degree in mechanical engineering from the University of Wisconsin, I went to work for a medical software company where I spent a lot of time helping customers track down bugs in the software. I later received a PhD in Mathematical Sciences at Carnegie Mellon on work combining probability theory and computation. After my postdoctoral work at Penn State, I decided to leave academia and pursue data science. It is a great fit for my background and allows me to solve challenging problems with real-world data.”

What is your favorite thing about being a data scientist? 

“I really enjoy problems that don’t have an immediate solution. In data science, one has to dig into the data, write code, make plots, ask questions, and just ponder. If one is fortunate, after all this work, the picture starts to clear up and one can begin to see what is really going on. It is this stage that I love. It’s the joy of discovery and getting to share that discovery with others.”

What is one thing that would surprise people about your field of work in data science? 

“Data science is much more than fancy algorithms and advanced tools. The majority of good data science is taking the time to carefully look at the data and understand what is going on. It’s great to do the fancy stuff but you also need to take the time to explore and think about simple plots and regressions.” 

What kinds of things do you think we should be doing to encourage more people to pursue a career in data science? 

“We need to encourage more underrepresented groups to explore data science and the STEM fields. While it’s important to learn the foundation of math, statistics, and programming, we should also emphasize the joy of discovery, exploration, and creating something new. For anyone interested in data science, I recommend that they find a topic that excites them and a corresponding data set. (There are so many free datasets out there.)  They should then take the time to explore the data, learning the tools as they go. (You know you are doing it right when you get so engrossed in your project that you forget to eat dinner.)”

What are your primary responsibilities at CIBO? 

“When I first joined CIBO, I helped code and design statistical algorithms. My role then transitioned as I led the project to infer management practices for our crop model. Now I manage the development of our proprietary productivity score and I’m responsible for producing our required deliverables on time. I work on both designing and analyzing our algorithm as well as coordinating with other teams to make sure we have all the needed data.”

What do you find the most rewarding about your work here?

“CIBO has a strong culture of collaboration and teamwork. I get to work with a number of very smart and talented individuals who are all willing to take the time to help each other out. Unlike data science roles at other Internet companies, my work is not focused on increasing marketing demand or click-through rates. Instead, we work on creating useful tools, such as our productivity scores, which increase transparency in the agricultural marketplace. It’s exciting to be at a startup where everything is new and moving fast, and where each of us is responsible for owning and executing on a significant portion of the company’s mission.”

About Jason Rute

Jason Rute is a Lead Data Scientist at CIBO, a science-driven software startup. He holds a Bachelor of Science in Mechanical Engineering, Mathematics, and Philosophy from the University of Wisconsin-Madison, along with a Ph.D. in Mathematical Sciences from Carnegie Mellon University. 

What led you to Computer Vision Engineering? 

Computer vision is an exciting field to work in. I love seeing my favorite kinds of math come to life in my day-to-day work, and it’s gratifying to have scientific insights that ultimately influence a product that real people will use.

What is your favorite thing about being a Computer Vision Engineer? 

I’m grateful for the opportunity to constantly learn new things. Computer vision is a fast-developing field, and I feel privileged that staying abreast of exciting new ideas and technologies is part of my job.

What brought you to CIBO? 

For a tech company, CIBO has a lot of scientists. I was excited by the opportunity to work on a product that is both scientifically principled and concretely useful.

What are your primary responsibilities at CIBO? 

I work with the modeling team. My day-to-day work ranges from working on the remote sensing platform to applying deep learning to satellite images.

What do you find the most rewarding about your work here? 

Collaborating with my team to build something cool. It’s a joy coming into work every day when your coworkers are friendly, knowledgeable, and dedicated.

About Ethan Sargent

Ethan Sargent is a Computer Vision Engineer at CIBO, a science-driven software startup. He holds a Bachelor of Science in Mathematics from Harvey Mudd College.