bockris

Do you mean applications? Making sure they are easily containerized and follow 12 factor guidelines is all you really need imo You should be able to run minikube on whatever dev machine you are already using.

I would like a tool that does something similar to this but on a more ad-hoc basis. i.e. to track my own explorations and understanding for fixing a given problem in a code base.

I'll also admit that I had a hard time understanding exactly how it worked.

It wasn't completely clear for each 'slide' what code was being referred to, I think the highlighting of the code being explained needs to be much more defined or the dimming of surrounding code needs to be more pronounced.

I think it would also be better if there were markers directly in the code side of the window that you could hover over to highlight parts of the slide that are relevant. It would also be beneficial if it also worked the opposite way. i.e. when clicking on a marker in the slide, scroll to the relevant code.

I also think your datapipeline example doesn't really show the value of the tool. i.e. it seems like a very straightforward piece of code that could just be read directly.

How are you planning to keep the slide information in sync with the files on github as they change?

I may not be the target audience for this because I would like it to work with code on my local machine and would only publish if I wanted advice from another programmer and then I would push up my branch and provide a link to the codeflow.

presumably your intended existing route was already on the main roads? while in navigation mode you can hit the search icon (magnifying glass) to search along the existing route. it won't be perfect but it should work for your use case.

this tool is fantastic. I have a custom written script to unload pictures/movies from an SD card and it copies them to a date based directory structure.

I added this right into that workflow and now newly downloaded pictures are immediately backed up to a google storage nearline bucket. I haven't done it yet but I could also target my home NAS so I would have our photo archive in 3 places.

common crawl

This data is mostly (completely?) based on personal usage.

  chrome://predictors/

They are completely worthless for the reasons you stated. A few months ago I connected with a colleague from ~ 15 years ago. He endorsed me for skills that weren't even invented when we worked together.

Recommendations are even more unreliable.

Case in point: A manager who was on the same level as my direct manager contracted an 'Architect' to come up with a high level estimate for a big project. The guy basically drew pictures in Enterprise Architect for 6 months and barely interviewed any stakeholders or technical folks. I was in the meeting where it was decided that it was a worthless exercise and big waste of time/money. We let him go right after he turned in the project plan and scrambled to redo the work. The manager that hired him wrote him a very positive, over the top, gushing recommendation on LinkedIn. I don't know if it was 'quid pro quo' or just the manager trying to save face but it definitely showed me that they are worthless.