Week 3 Student Handout

One-page take-home for the Course 3 session. Print landscape and bring it with you, or save as PDF.

Print this page in landscape from Chrome or Edge: Ctrl+P → Layout Landscape → Margins None or Default → Background graphics on. The site chrome above is hidden in print.

In the print dialog, leave Margins at Default (or pick None). Picking Custom or Minimum shrinks the printable area below what this layout assumes and the handout tips onto a second page.

Download PDF Pre-rendered Letter-landscape one-pager — for shared / kiosk laptops where the print dialog is locked down.
3 Course 3 · Student Handout

Platform Training

Three deployed Power Platform builds in four hours — centaur, cyborg, and your own problem.

Duration4 hr (2 breaks)
AudienceBuilders
PrereqCourse 2 + M365
0:00–0:30M1 Setup & Mode Refresher30 min · talk
0:30–1:30M2 Build #1 · Centaur60 min · build
1:30–1:45Break + Failure share15 min
1:45–2:45M3 Build #2 · Cyborg60 min · build
2:45–2:55Break10 min
2:55–3:55M4 Build #3 · Your Problem60 min · build
3:55–4:00M5 Frontier Map & Wrap5 min

Bring with youDone before you walk in

  • Course 2 prototype with notes on what worked, what didn’t, and one failure case. Failure-sharing is the first 10 minutes after Build #1.
  • Power Platform access verified. Sign in to Power Apps, SharePoint, and Power BI on your laptop before the session. Test creating a one-row list.
  • AI tools. GenAI.mil or CamoGPT for any CUI work. ChatGPT / Gemini OK for non-CUI exercises.
  • Build #3 problem statement ready in your head: one sentence on tool + user. You will declare it out loud at 2:55.
  • Your frontier sticky. You will add a real, specific row to it during Module 5.

Key termsThe vocabulary you’ll hear today

  • Phase boundary (centaur)A non-negotiable checkpoint between design, build, and verify. You cross only when the previous phase is complete and signed.
  • Whiteboard firstEvery business rule, data flow, and edge case written down before AI runs a single prompt. Centaur opener.
  • Cyborg pivotA mid-build moment when you throw out the data model AI helped you start with and rebuild — on purpose. Discovery in action.
  • The declarationBuild #3 only: state out loud the tool, the user, the mode (with reason), and two specific frontier risks before you open the platform.
  • Frontier recognitionCatching a wrong AI answer fast enough to course-correct in the same prompt — not three prompts later.
  • Deliberate failureA point in the build where AI is known to give wrong instructions. You let it happen so you learn to spot it.

Exercises in classThree deployed builds — what each asks of you

Build #1 · Centaur (60 min) — Request Routing. Submit → auto-route by dollar amount → approver decides → requester notified. SharePoint list + Power App form + Power Automate flow. Phase 1 (10 min) is whiteboard only — no AI. Done: $400, $1,200, and $3,000 test requests each route correctly and the right notification fires. Bonus: 48-hr approver timeout escalates.

Failure share (10 min, post-Build #1). Bring one failure case from your Course 2 prototype to the room. We capture the patterns — specific is better than tidy.

Build #2 · Cyborg (60 min) — Training Tracker. “See which Marines have completed which training.” SharePoint list, input form, Power BI dashboard with conditional formatting, calculated days-until-due. You will restructure the data model mid-build — that is the point. Done: a dashboard that colour-codes status across five training events, and you can explain why you reshaped the data.

Build #3 · Your problem (60 min) — Declare first. One sentence: tool type + user. One mode + why. Two specific frontier risks. If you can’t name two specific risks, revise before you start. Done at 3:45: the simplest version of your tool runs against real input, and you can defend the mode choice.

M5 · Frontier map update (5 min). Add one new row to your unit’s frontier map from today. Done: one specific task, one specific failure mode, one specific re-test trigger.

Anchor phraseThe boundary between you and the AI is a tool. Pick the right one for the problem.

What you’ll be able to doBy the end of the session

  • Build a Power Platform tool in centaur mode with verification at every phase boundary.
  • Build a Power Platform tool in cyborg mode, including a deliberate data-model pivot.
  • Choose between modes for your own problem, defend the choice, and ship in 60 minutes.
  • Recognise a wrong AI answer fast enough to course-correct in the same prompt.
  • Add a real, specific row to your unit’s frontier map.

HomeworkRequired to attend Week 4 Advanced Workshop

  • Finish Build #3 to deployable state — users can open it and complete the core action.
  • Run it through the EDD SOP QA process. Don’t skip the verification checkpoints — they are the entry ticket.
  • Document three specific failure cases you encountered while finishing. Log them on the shared frontier-map doc.
  • Identify one capability surprise — somewhere AI performed better or worse than you expected.
  • If you get blocked, contact the program lead before Week 4. We’d rather unblock you than have you skip the workshop.