I am a student at South Puget Sound Community College, and I am asking for something that may look unusual on paper: permission to complete my mathematics coursework as a self-paced independent study, joined to a parallel path in data science. I want to make the case for it here in full, because I believe that once the design is laid out, it will look less like an exception to the way a college works and more like an expression of what a college is for. The proposal asks for flexibility in pacing and format. It does not ask for any flexibility in standards. Every claim I make about what I have learned will be checked on campus, by the department, on the department's terms.
I took mathematics as part of my undergraduate program, and my background since then has been in psychology research and health. I'm now looking at master's programs in data science and artificial intelligence, and I want to refresh all of my math before I start one rather than build on half-remembered foundations. The fields I want to work in all run on that foundation: philosophy, experimental psychology, experimental philosophy with data analysis, health research, and biomedical engineering. They need everything from the logic and formal reasoning philosophy asks for to the statistics, linear algebra and calculus that data work runs on. This plan is how I intend to rebuild it properly, from the ground up.
The heart of the proposal is a sequence of thirteen blocks spread across 116 weeks. It begins on November 2, 2026 and ends around January 21, 2029. I plan to work fifteen to twenty hours every week, and I have built buffer time into the schedule on purpose. Learning mathematics is not a straight line. Some weeks a concept will click on the first evening, and other weeks a single idea will take far longer than I expected. A schedule that pretends otherwise breaks the first time it meets reality. A schedule with room to breathe can absorb a hard week without abandoning the destination.
The order of the blocks follows the logic of the subject itself. I start where the foundations are laid, with intermediate and college algebra, and then move through matrix algebra, geometry, trigonometry and precalculus. Counting and probability lead naturally into statistics, and from there the path climbs through three blocks of calculus before finishing with linear algebra and differential equations. Each block prepares the ground for the one after it, so nothing arrives before the ideas it depends on.
| Block | Topic | Length (weeks) | Plan weeks |
|---|---|---|---|
| 1 | Intermediate algebra | 16 | 1–16 |
| 2 | College algebra | 8 | 17–24 |
| 3 | Matrix algebra | 4 | 25–28 |
| 4 | Geometry | 8 | 29–36 |
| 5 | Trigonometry | 8 | 37–44 |
| 6 | Precalculus | 6 | 45–50 |
| 7 | Counting & probability | 4 | 51–54 |
| 8 | Statistics | 8 | 55–62 |
| 9 | Calculus 1 | 12 | 63–74 |
| 10 | Calculus 2 | 12 | 75–86 |
| 11 | Calculus 3 | 10 | 87–96 |
| 12 | Linear algebra | 10 | 97–106 |
| 13 | Differential equations | 10 | 107–116 |
The full week-by-week version lives in my math plan, and the materials, tools and records that support it are collected in my Learning Hub. Both are public, and I invite anyone weighing this proposal to look through them.
I am not proposing a math course with some programming bolted onto the side. I am proposing a single, coherent curriculum in which mathematics and data science are two faces of the same work. Every topic in the sequence includes Python exercises and hands-on work with data. When I learn how a function behaves, I will also write code that computes it, plots it and tests it. When I study probability and statistics, I will not just solve problems on paper; I will work with actual data and see the ideas hold up, or fail to, in practice.
The reasoning behind this pairing is simple. The mathematics is the foundation. It is what lets me understand why a method works, when it can be trusted, and what to do when it breaks. The Python and data skills are the marketable outcome, the part of the work that turns understanding into something I can build, share and be hired for. Together they make a record that should mean something both to employers and to the universities I hope to transfer to. A transcript tells a reader that I passed a course. A body of documented, working code shows what I can actually do with what I learned.
A self-paced path only works if it rests on materials that are rigorous, complete and available to me at every hour of the day. I have built the curriculum on free, openly licensed textbooks. From OpenStax I will use Intermediate Algebra 2e, College Algebra 2e (including its seventh chapter for matrices), Precalculus 2e, and Calculus Volumes 1 through 3, all released under CC BY-NC-SA 4.0, along with OpenStax Statistics under CC BY 4.0. For the final stretch I will use Jim Hefferon's Linear Algebra and Jiří Lebl's Notes on Diffy Qs. None of these texts costs me anything, and every one of them can be read and checked by any faculty member who wants to see exactly what I am studying.
Around those texts sits a set of tools I will use deliberately rather than casually. Khan Academy and Brilliant give me practice and alternate explanations. Wolfram Alpha lets me check computations, while GeoGebra and Desmos let me see the geometry and the graphs behind the symbols. Google Classroom gives the work a place to live, and edX, Coursera and LinkedIn Learning offer structured courses when I need a second voice on a topic. Each week I will produce a Jupyter notebook that ties the mathematics and the code together in one place, so that my progress is not a feeling but a document.
I think of these tools less as separate apps than as a small team of AI agents with one shared purpose: helping me learn the mathematics and prove that I have. One drafts lesson plans, another quizzes me, another keeps the logs and reminders. With the college's permission, agents like these could also work with university services on a student's behalf, for example checking deadlines, booking a proctored test or organizing materials from the course site. I've already run into this: when I tried to connect my agent to the college's email to gather some information, IT replied that it wasn't supported. That's fair, and it's exactly the kind of integration that would need the college's permission and support. These agents would always act under the student's name and within the college's rules, never around them.
I intend to use artificial intelligence openly and on purpose, in two roles, an approach that research on tutoring software supports (Kulik & Fletcher, 2016; Ma et al., 2014). The first is as a coach: something I can ask to explain an idea another way, to walk through where my reasoning went wrong, or to suggest the next problem that will stretch me. The second, and to my mind the more important, is as a test. I will use AI to question me, to probe whether I really understand a concept or have only memorized a procedure, and to push on the places where my understanding is thin.
I will keep daily logs of this work. Those logs record what I studied, what I asked, where I struggled and what I changed as a result, along with my own feedback on what is working and what is not. That feedback loop matters. A self-paced plan succeeds or fails on whether the learner notices problems early and adjusts, and a daily log makes that noticing a habit rather than a hope.
I know that many faculty members have not had much training in AI tools, and I do not think it would be fair to ask anyone to learn a new set of tools just to supervise me. So I am offering to carry that weight myself. I will provide screen-share demonstrations that show how I set up and use each tool, so that nothing about my process is a black box. The faculty role is simply to review the work. I want this arrangement to feel like a service I bring to the department, not a burden I place on it.