Check-in meeting instructions

As we approach the end of week 2 and look ahead to our final presentations during week 3, we’d like to schedule a slightly-more-formal-yet-still-pretty-informal 1-on-1 check-in with each of you to make sure that you are in a good position to finish out the module strong. These check-in meetings are scheduled for 20 min, but they can take less or more time, as scheduling allows and as you need.

We see these check-ins as serving a few purposes:

  • Organizing your thoughts and summarizing what you have learned in the first week of self-directed research.

  • Getting your questions and concerns addressed before the last week.

  • Giving you practice articulating your thoughts and making slides for your final presentation.

Summary slides

To help focus the conversation, we ask that you prepare the following for this meeting: a super short presentation with no more than two slides. The following are suggested contents, that you can also merge onto a single slide if you want.

  • The first slide could summarize some things you learned, tried, and/or found interesting so far. Please include at least one figure of your data/models/other.

  • The second slide could list questions you may have going forward, a figure with a poor model, etc.

Attention

Please don’t let this check-in meeting stress you out! The intent is for the exact opposite! 😅

Again, this meeting is one of several small checkpoints (that are admittedly in rapid-fire succession) to give you tangible milestones to work towards and keep things flowing. 🌊 We understand this is still early in a typical research process, so to clarify our expectations for what you should have prepared for this check-in, here are some examples:

  • You can produce a data visualization of some interesting (or not interesting) trends that you observed during exploratory data analysis of your feature space. We could then have a discussion about what other analyses to do, ML models to build, etc.

  • Perhaps you’ve already trained and validated ML models using the features you selected. Try to put together a figure and/or table summarizing the model performance, and we can discuss how to improve the model, what things to try next, etc.

  • Perhaps you got super stuck somewhere along the way, or found some interesting physics/chemistry but aren’t sure how to proceed with a programmatic, data-driven solution. That’s OK! Teach Enze about what you ran into and let’s see if we can get unstuck.