Last week was focused on refining project ideas. When I took this course last semester, I was taking an existing project to the next level, so this planning stage was easier. This semester, I’m starting a fresh project, so there are inevitably more unknowns. However, I think my experience doing my project last semester and my experience with many of the technologies I plan to use this semester will allow me to create an effective and reasonable plan for my project.
More concretely, I have decided on doing a reverse proxy project and have been working on defining my specific goals and assessments. A reverse proxy seems particularly flexible and suitable for this kind of timeline, since there are many different functionalities that can be added or removed from the plan, still resulting in a solid, complete project.
This week, I will complete the project proposal. I would say most of the core planning of the project is done at this point, but I still need to compose my ideas in narrative form and flesh out the other sections. I’m confident that I will be able to continue to refine my ideas and set myself up for a productive and meaningful semester.
Since I’m planning to apply to jobs in Japan, one impediment I’ve been feeling is my incomplete sense of the job market there. Of course, I don’t have a complete sense of the job market in the US either, but factors like language and fewer contacts do make things a bit harder to grasp.
This will be something I continue to work on over the course of the semester, and I don’t think I should expect my choice of project to be some perfect key to me getting a job. I think I’ve found a good balance between career relevance and my own interests with what I’ve planned so far.
As I’m writing this, I just got the notification that I passed JLPT N1 (the highest level of the Japanese language exam for nonnative speakers), so that’s at least one major impediment out of the way!
The primary approach I took last week was gathering more information on potential project ideas and my eventual topic of choice via conversations with Claude. On the employment landscape front, I also spent some time going through job boards.
While, of course, one can’t just trust everything an LLM says, I feel that I’ve made significant progress in being able to determine what kinds of answers are likely to be unambiguous from training data and what kinds of answers are prone to be unreliable or outdated. For something as exploratory and wide open as planning out this project, more traditional tutorial-oriented resources never quite seem to fit with my thought process.
I would say my process last week was effective. Once I get into the hands-on implementation, the task will be different enough to require a process that looks very different. However, overarching strategies for optimizing the pace and depth of my learning using AI will remain relevant.