The Artifact

<aside> 💡

While this doc is an accurate representation of why we are building the automated feedback workflow, we made a couple changes:

• We have two DBs:
  • calendly_meetings: Stores the scraped calendly meetings for any analytics in a postgres DB (mostly a denormalized schema to make building the autoemailer easier)
  • feedback_form_responses : Stores the feedback form responses </aside>

The workflows:

• calendly_ingestor: Each day, scrape the calendly API to get the meetings within the past 30 days and write to calendly_meetings
• feedback_form_autoemailer: Each day after calendly_ingestor workflow finishes (we have one hour time span so it should be plenty of time), look for the meetings where feedback form email wasn’t sent and call Retool Email to send the feedback form email

Retool Feedback Form:

• https://datasciencegt.retool.com/form/2264de5b-9cf1-4706-8651-0a068b1ff734

Analytics Tool (internal facing):

• https://datasciencegt.retool.com/apps/430be3ee-3c29-11f0-b26a-477f84695787/Analytics Portal/page1

Rough architecture diagram of calendly_ingestor + Feedback Form Autoemailer

Rough architecture diagram of feedback form backend

Background/Context

I currently help students looking to break into software engineering by making time available for 1:1 conversations with me. Currently, students can schedule 1:1 consulting and mock technical interview sessions through my dedicated Calendly link. After each session, feedback is collected manually by sending a Google Form link to the attendee. This process is time-consuming and prone to inconsistencies or delays.