Skip to content
Clawdemy

References: AI on a real project

Source material

Section titled “Source material”

This lesson is an original adaptation. Its method (name the audience, platform, goal for every piece and vary the shape), the ask-me-what-you-need move, the persona rehearsal with its tunable pressure, the classroom’s split verdict on simulation and the teacher’s caution that the machine coach is a complement to human preparation rather than a replacement, the spectrum from comfortable uses to hard-to-defend ones, and the transparency remedy come from the seventh class of a Harvard Kennedy School course. The job-search scenario, the food-pantry variant, and the prose are our own. Clawdemy is independent of Harvard, which has not reviewed or endorsed this track.

  • The Science and Implications of Generative AI (HKS DPI-681M), Harvard Kennedy School, Spring 2024. Faculty: Sharad Goel, Dan Levy, and Teddy Svoronos. This lesson adapts Class 7 from the Spring 2024 course site, whose content is licensed under Creative Commons Attribution 4.0. The class is the case study that closes the course’s unit on using generative AI: everything the unit taught, run end to end on a single communications campaign. The lesson’s two in-marks quotes, “please ask me what you need to know to do this task well” and “closer to what you do in your day-to-day life,” are the session’s teacher’s words from this class’s lecture videos.
  • Official course lecture playlist on YouTube, Harvard Kennedy School. The full lectures, free to watch. We mean it when we encourage you to take the original course alongside this track: good teachers deserve more students.
  • Provenance note: quotations and classroom details in this lesson were drawn from the transcripts of the official Class 7 lecture videos in the playlist above, obtained from the official Harvard sources only. No third-party re-uploads or mirrors were used.

The examples behind the lesson

Section titled “The examples behind the lesson”

The lesson’s project is our own scenario, so it leans on one external study, cited here with its limits stated.

  • Eye-Tracking Study (PDF), Ladders, Inc., 2018. The recruiter eye-tracking study behind the opening’s claim that a recruiter with dozens of applications waiting may give it less than a minute. Its limits are why the lesson hedges: the study was run by a jobs marketplace, not independent researchers; it dates to 2018; its eye-tracking sample was small; and its methodology has been criticized. So the lesson keeps the hedged form rather than quoting a precise seconds figure. The durable, directional point is simply that initial resume review is fast. For a press account, see HR Dive’s coverage of the study, 2018.
  • The opening’s companion claim, that the resume was quite possibly filtered through screening software before a human saw it, is hedged for the same reason: screening practices vary widely by employer, industry, and tool, and the lesson asserts only that filtering is common enough to write for.

Going deeper

Section titled “Going deeper”

On this site

Section titled “On this site”
  • Asking well: the anatomy of a good prompt, lesson 2 of this track. The prompt anatomy (task, instructions, context) this lesson runs at speed, and the home of the persona move the rehearsal carries.
  • Beyond the chat window: tailoring AI to your work, lesson 3 of this track. The tailored assistant, standing instructions plus your documents, that this lesson reuses as its working vehicle.
  • Should AI do this task?, lesson 4 of this track. The two filters this lesson applies to a live project, and the canonical home of the privacy question asked before every paste.

Adjacent topics

Section titled “Adjacent topics”
  • This lesson closes the track’s first movement. The next lesson opens the second, from how do I use this well to what does all this mean for my world, starting with the kinds of risk this technology carries and the levers for managing them.