All of a sudden, three months in Mali are coming to a close. In some ways, I’m leaving with my project still in its beginning stages. We are far from adding a new product to the catalog. But at the same time, it is rewarding to know that I was able to drive a project from an abstract idea all the way to a tested, fabricated, and promising prototype. As I wrap-up, I’ve been reflecting on the ways my work, in its different phases, existed within the context of the overall organization. In some instances, I felt the need to create new infrastructure but in others I was able to seamlessly leverage existing processes. Together, these contrasting experiences kept me on my toes and taught me a lot about myAgro as an organization.
Starting with instances where I felt like I was treading new water, R&D as a concept is new to the tools team. From establishing a prototyping process without any in-house prototyping capacity to designing an efficient testing process for one person to execute, I figured out a lot by trial and error. Any time we needed something built, I drew simple sketches and used photos to communicate with a fabricant in the industrial market. Much to my surprise the first time, I was impressed by how well our contact at the market understood my ideas. But sometimes, if I wasn’t perfectly explicit, my intentions would be lost and we’d need to try again. On a project management front, it was definitely an added challenge to have prototyping outsourced, from planning a tightly budgeted calendar to bargaining the price each time, but on a whole, we worked through the cyclical design process together. On the one hand, I have a lot of appreciation for being able to prototype in-house. The ability to make small adjustments and refine a design in parallel with the testing process is very convenient. But on the other hand, the mechanical expertise of fabricants was highly useful in building fully functional prototypes from scratch, quickly. For the testing, it was important to me to streamline and identify the highest priority tests to ensure Nouhoum wasn’t unnecessarily performing tedious tasks. But despite these intentions, it still took a lot of tests to investigate the effect of various variables on the performance of the prototype and identify the best possible combination. I tried to spend time showing Nouhoum how the data he had collected was leading me to make different decisions, especially when I could capture his results in a meaningful graph, but this entire process was very new to him. It was hard to gauge how much he was able to connect the purpose of the tests to the overall mission of the project. Still, it was an exciting first step for him to live and breathe the design process myAgro presents as our guiding philosophy. Initially, I was a bit startled by the task of taking on this project without there being any precedent for hands-on development within the team, but by the end I came to appreciate how it made me think more carefully about how to start making the process accessible, for engineers and non-engineers alike.
While a lot of this project felt separate from the existing programs or projects within myAgro, the last few steps integrated solidly. After running dozens of tests with the semoir to measure how accurately our prototype could drop seeds with the correct spacing, the more thorough way to test its efficacy was to use it to plant, and then monitor germination. To do this, I worked with the agricultural team and found they have impressively thorough protocol in place for running different kinds of randomized control trials. All I had to do was write up the details of our test, and the rest fell into place. On the day of the test, measured and labeled plots of land, a couple laborers, and a donkey awaited. And now, even without my overseeing, these plots of land are being nurtured and observed based on my criteria. For the first seven days, they even hire a bird-chaser to arrive early in the morning and stay late in the evening, to ensure the birds don’t snack on the newly dropped corn kernels. I laughed when I heard this, both because I could never have known this was necessary and because the image of hiring someone to chase birds seemed absurd. But apparently it’s the most effective way to ensure the trials are uninterrupted, so I went along. The beauty of being able to hand off this test to the agricultural team is that even though I’m leaving, these plots will be maintained and studied through harvest. And the results will show how the myAgro semoir, our new semoir, and careful hand planting all compare. Following this trial, my project ended with farmer demonstrations and feedback sessions. These insights are fundamental to the success of the project overall and could never have been gathered on my own. Thankfully, myAgro’s extensive field team was there to support me and facilitate. A field coordinator organized a group of farmers to meet with Nouhoum and myself, and we had the chance to ask questions, hear concerns, and gauge interest. Encouragingly, reactions were positive and the concrete steps being shown on our end to lower the price of the semoir were acknowledged and appreciated. One farmer told me he would mobilize the entire village to buy semoirs if I promised to lower the price. I laughed and told him we should sign a contract. But jokes aside, the underlying sentiment was reassuring. After spending a lot of time more focused on the success of the prototype from functionality, agricultural, and financial perspectives, it was rewarding to connect with users and see how the prototype would fare in their eyes. Ending this project at a time when the results look promising and other teams are getting involved makes it harder for me to detach, but also signals to me that my project has a chance to take root in the organization.
Quite separate from my main project, I ended up working on a side project, writing excel macros and functions to help process the vast amounts of data that need to be compiled into reports each week. It all started when I heard how much time people spend weekly, manually manipulating excel files in the same, repetitive way. I’m no excel expert, but I hadn’t done much coding in a while and the puzzle of teaching myself a new coding language while also making my colleagues’ lives easier drew me in. Through this process, I learned how every week, new discussions led to new kinds of reports, resulting in new or modified macros. Process was being built on the fly. In this case, I found it gratifying to know I was helping and contributing a unique skillset to the team, but it was hard to create a system that could properly address all the desired steps without anyone knowing what the next step would be ahead of time. In my last week, my work had generated enough interest to warrant me running a small introduction to macros, but it was more an introduction to the idea that weekly reports don’t need to be so tedious and less of a lesson on how to code. I hope that some of what I leave behind continues to be useful, but without knowing how dramatically needs might change week to week, it’s hard to know. Regardless, it was exciting to contribute in a way I had never anticipated. I joined myAgro technically as a mechanical engineer, I self-define as more of a design engineer, I have the most experience with electrical/digital engineering, and I ended up applying some coding skills semi-accidentally. I continue to appreciate how my background has given me the flexibility to apply myself in such a diversity of ways.
It’s always hard to leave a project and not know if and how it will be carried forward. In places where my work plugged into well established and on-going operations, it’s easier to have confidence that that phase of the project will be carried out. In places where my work felt really new, floating, or somehow detached, it’s harder. Still, I’m optimistic. Given the director’s enthusiasm for continuing a focus on tools, engineering, and R&D and given conversations around what new R&D work is in the pipeline for the next year, especially around continuing the semoir redesign, I hope to see myAgro grow in these areas that felt small and budding during my time here. The awaited results of the trial play a large role in determining the feasibility of my design specifically, but regardless of what any final product looks like, I’m glad to have been part of kicking off this process.
Lots of photos of semoirs!
The title photo and the first row are photos from the industrial market where we had prototypes made. I found it fascinating to see how small scale manufacturing occurs.
In the row below, the left photo shows one of the early prototypes with the myAgro semoir. And the right photo shows our last prototype.
The next row of three photos shows Nouhoum conducting tests.
The row below is from a trip I took to Sikasso and met with farmers. The back of my shirt says “nothing is impossible with myAgro” in French.
The next three rows are from the final trial.
The first shows the three methods – hand planting, myAgro semoir, and the new prototype.
The next two rows are photos I liked of the donkey, the planting, and the team that helped implement the trial. One of the laborers said he liked our new prototype more than the myAgro semoir so he was excited.
The last row shows photos of Nouhoum leading a farmer demonstration. Even kids gathered around, but they ran away when I jokingly asked if they would buy a semoir too.
