By Jonathan Briggs, Director of Strategy, Technology, and Innovation
Since its founding, Eastside Prep has valued project work. This is evident in the design of our class experiences as well as our clubs and academic teams. In our short history, examples of project work at EPS include impact projects, project weeks, and senior projects. All of those efforts had great successes and as is true with any new creation, opportunities for improvement. Those three examples no longer exist in their original form because of something else that Eastside Prep values—constant refinement. Instead, elements of those programs evolved into pieces of current programs such as EBC Week, maker spaces, Independent Studies, and Seminars.
Over the years, we have noticed that Independent Studies exist on a spectrum between academic research on one side and building a product on the other. The product-focused independent studies would often have to contort themselves to fit into that mold. For example, we have had students direct plays, write mobile apps, script novellas, and produce record albums. In winter 2022-23, we started prototyping what a project-focused Independent Study might look like with seven willing students.
In an academic or research-focused Independent Study, the area of study tends to be well known. Often, resources are organized into topics and as you accrue resources, you have a general idea as to how long it will take to learn each topic. Certainly, there are tangential things that you may need to research as you go that will slow you down, but they tend to be minimal. The level of depth of research can also be adjusted to keep the student on pace. It is reasonable to assume, for example, that a student could pursue an Independent Study in quantum mechanics over a trimester even though people get PhDs in the same topic.
In a project, the outcome is easy to define but the process, and time, to get there is far more difficult to predict. Large companies routinely miss product deadlines either in time or in feature set despite years of dedicated effort to do otherwise. This has led to some of my favorite sayings, such as, “The first 80% of your project takes 80% of the time, the last 20% of the project takes the other 80% of your time.” Or “The project will take twice as long as you think it will, even if you account for this rule.” To that end, we are trying a different format for independent projects. Our hope is both that this works well and that we have continued to evolve the format between when this was written to when it arrives in your mailbox.
Independent projects are currently set up in the following way. Students focus on mapping the first half of the term’s work on the project with specific resources and deadlines that they would have to meet in order to get a proof of concept done by mid-term. What a proof of concept looks like is quite project specific. It could be an application that works but has had no attention paid to the user interface or it could be the inverse, a user interface without any working application behind it. Or it could be an outline for a novel or one polished completed chapter. The goals of the first part of the term being experiencing the actual pace of a project compared to the planned pace, in addition to building their knowledge around the problem or processes involved. At mid-term, students then formulate an informed plan for the second half of the term, knowing that they will have to present their final product to the community. This meta understanding is supported through weekly status reports that the student submits in addition to scheduled and unscheduled meetings with their mentor.
So far, the results have been encouraging. We have seven students engaging in independent project work in a variety of domains that address their individual curiosities.
- Exploring Compliant Mechanisms: These are mechanisms that use the material’s properties to provide a level of movement. Interestingly, they are used in very inexpensive devices, like bending plastic to make a cheap hinge, and very expensive devices, like extremely precise aligning of mirrors in space telescopes. Grant (’23) is studying, designing, and producing these mechanisms.
- Generative Adversarial Networks: These are AI networks where one AI system tries to build data that another system will accept. Over time, the building network learns how to produce the most convincing data. This technique is how researchers create images of people who don’t exist. Zubin (’23) is building image- and text-based GANs and is curious if there is a way to merge the two.
- Fabrication with CAD and CNC: Modern machining uses a variety of computer tools. CAD, computer assisted design, builds models of objects. CAM, computer assisted machining, translates those models into paths for a physical tool to take. And CNC, computer numerical control, takes those paths and cuts the part. EPS is fortunate to have all of the pieces of this process available to students.
Evan (’23) is working to understand that process from idea to completed part as he builds original metal parts, some of which will be used in the EPS electric vehicle project. - Visual Representation of Reinforcement Learning: Reinforcement learning is how an artificial intelligence system gets trained to solve a problem. The process feels a bit magical and that bothered Jules (’23). He has set out to build visuals that can give someone an intuitive sense of how that works. In particular, he wants it to use a real AI network’s data, not be a conceptual animation.
- Stock Market Prediction: On the surface, it seems that there is so much information out there that stock market prediction should be possible. There have been many efforts over the years with mixed results. What is certain is that it is a great place to try to apply machine learning tools. In doing so, you will understand how the tools work and what their limitations are. Eric (’23) is doing just that during winter term.
Human-Centered Design: Designing products for people’s bodies is a particular challenge. Size, space, and aesthetic considerations are ever present. While initially focused on building prototypes with our CNC wood router (that spent much of winter being moved), the project has shifted focus toward what it takes to patent a design. For Stan (’23), this has been a lesson in adapting to circumstances and maintaining forward progress regardless.- Rocket Nozzle Design: The interest in space and particularly rocketry has increased substantially over the last ten years. Understanding how air and fuel are mixed, ignited, and directed is a complicated undertaking, and Fino (’23) is working toward building a testbed for different nozzle designs
having to learn some chemistry and thermodynamics along the way.
While our initial term has been projects in the technology discipline, we are excited to see how independent projects manifest themselves in other disciplines starting in the spring with Fine and Performing Arts. EPS students have always pursued their curiosity and learned by doing. The Independent Project serves as a mechanism to amplify those efforts.