Learn
By the end of this unit
- Construct a usable no-code route from permitted input through an explicit AI step and human check to an output or handoff.
- Test varied inputs, record defects, and make a traceable revision.
Planned guided work including the short textbook read-and-apply: 77–127 minutes, plus other project work. This is a design estimate, not measured learner time.
How to read this lesson
The labels describe how to use a passage. Color does not prove a claim is true or that a response is correct.
- Explain — Understand the idea.
- Try / predict — Make a choice or prediction before the reveal.
- Evidence — Inspect the named source or supplied data.
- Worked reasoning — Compare the steps after your own attempt.
- Caution — Check a limit, assumption or unsupported conclusion.
Each figure has its own local legend for the data. Blue selection outlines mark the current selection; other outlines are explained by each figure’s local legend. Selection is not a correctness judgment.
Construct ReplyKit, one usable part at a time
A product can be a set of dependable working documents. It becomes usable when another person knows what to enter, what to do, what result to expect, and when to ask for help. For this first version, a document editor and calculator are enough. A shared folder is convenient, but you can hand over a complete file without a new account.
- Create a Start page. Name ReplyKit version 1, its intended coordinator, the task, and its limits. Tell the user to read the source card before drafting.
- Copy the approved source card. Preserve S1–S6 and the version. Keep it beside the response, not hidden behind the author’s explanation.
- Make an intake table. Include case reference, customer question, relevant supplied facts, missing facts, and proposed policy source. Use fictional information.
- Make a draft-and-evidence table. Use columns for the proposed reply, material claim, source, and check result. Paste the bounded instruction from Unit 1 into the working file.
- Add the review checklist. Check source consistency, boundary conditions, unsupported promises, useful next step, and correct handoff. Provide approve/revise/hold choices.
- Add a delivery record. Record version, permitted recipient, actual delivery status, reviewer, and unresolved issue. Leave completion fields blank until an action occurs.
You have now built the service’s inputs, working process, quality control, and output. Test the complete route rather than spending all your time formatting its front page. A form, spreadsheet, or visual no-code tool can replace a document later if it makes the task easier; it does not change the evidence standard.
Make the package for your chosen capability
Choose the row relevant to your existing project; these are worked alternatives, not extra assignments. Every package still needs permitted input, a candidate, an inspectable check, a corrected useful output, and an exception route. The supplied examples are fictional.
| Capability | Permitted input and supplied candidate | Inspect and repair | Usable handoff and next test |
|---|---|---|---|
| Event extraction | Notice A: “Repair meetup, 12 May, Room 2.” Notice B: “Garden exchange, 15 May; location to follow.” A candidate table puts Room 2 in both rows. | Compare Event, Date and Location with each notice. Replace B's location with “Not supplied”; do not infer a year. Keep the original candidate and correction. | Provide notices, field rules, corrected table and a missing-field request. Give the reviewer a third notice with a missing date and observe whether the kit preserves the gap. |
| Image or poster preparation | An approved brief permits “Garden exchange, 15 May; venue to be confirmed.” A supplied poster candidate is described as displaying “16 May, Room 2” and a sponsor logo that was not provided. | Correct the date, show the venue as unconfirmed and remove the unsupported sponsor. Check all rendered text, reading order, contrast and permitted assets. This text-described candidate rehearses checking; it does not demonstrate image generation. | Provide the approved brief, permitted assets, corrected usable poster or accessible layout, alt text and a release checklist. Ask a reviewer to find the date and venue status. With an actual image tool, save its real candidate and inspect the image rather than trust its own description. |
| Voice or transcription assistance | An approved script says “Doors open at 6:15 p.m.; do not bring equipment.” A supplied transcript reads “6:50 p.m.; bring equipment.” | Correct the time and restore the negation. Compare names, numbers and instructions with the permitted source. A script/transcript comparison rehearses content checking; it does not measure speech-recognition accuracy. | Provide the source, corrected transcript/captions, uncertainty markers and a human handoff. Ask a reviewer to identify the opening time and equipment instruction. If permitted audio is available, test the actual clip, record assistance and retain only authorized information. |
For data analysis or prediction, use the same package with an input table, operation, independent calculation or baseline, evaluation and clearly conditional output. Adapt the three bench cases below to your selected capability: ordinary input, missing information and a consequential exception.
Worked ordinary input and completed output
Input: “I want to cancel. My workshop starts in 48 hours.” Source: S2. A supplied flawed draft promises a refund. The reviewer replaces it with: “Your request is within the 24-to-under 72-hour window, so the policy provides credit valid for 90 days. A coordinator must confirm the booking details and next step; this draft does not mean a credit has already been issued.”
Review result: Correct window; no unsupported refund; credit duration retained; no false completed action. Record: Approved classroom draft; no external delivery. If the time until the session were unknown, ask for it before selecting the rule.
See the decision
A repair is only as good as its next test
Predict which cases a new rule fixes—and which it leaves unresolved.
An original construction demonstration using S2. These supplied prototype outcomes are separate from the 12-case benchmark and from the actual test record you must collect.
| Version or reveal | Cases passing | What changed |
|---|---|---|
| Initial rule | 1/3 | Refund eligibility is incorrectly used for every input. |
| Require timing | 2/3 | Ask when the session starts if timing is missing. |
| Route late exceptions | 3/3 | Under 24 hours needs owner review. |
| Reveal the 24-hour boundary | 3/4 | The additional case requires credit, not a refund. |
| Add the credit window | 4/4 | Apply credit from 24 to under 72 hours. |
All three initial cases can pass while an untested boundary still fails. The denominator changes when the fourth case is added. Retest the earlier cases after each correction; four passing cases do not establish universal reliability.
Use it in your project: Choose a boundary or missing-input case for your own artifact, make a meaningful repair, and record the actual retest.
Required textbook reading and application · about 12 minutes
Read, then apply one idea
Read only the passages listed here. Mathematical derivations, algorithms, code, and end-of-chapter exercises are not assigned. The time estimate includes the brief passage and one connection to your existing project update; reading the passage alone usually takes much less time. It is a planning estimate, not a timer.
1.6.2: Tasks
David L. Poole and Alan K. Mackworth. Artificial Intelligence: Foundations of Computational Agents. 3rd edition. Cambridge University Press, 2023.
Read: Go to 1.6.2 Tasks. Read its opening paragraphs, Figure 1.9, and four-step list. Stop immediately after that list, before the discussion distinguishing long-term knowledge and immediate belief.
Why this unit: Connects a stated need to inputs, an output or action, and interpretation of that output.
Course connection: Figure 1.9, in words: define the task and what counts as a solution; represent it for a computer; obtain an output or action; then interpret that result in the original task. This is our text description of the central idea, not a copy of the figure. Build the smallest version that can carry out the selected task with approved sample input. A paper mockup can clarify an idea, but the course's final AI-assisted artifact must be usable and tested according to the project requirements. Follow the original no-code build guide for the chosen available tool; these readings explain the design logic, not tool-specific setup.
Apply one idea: Choose one prompt for the brief connection in your existing project update.
- Identify the input, the usable output, and the question your first prototype will test. State which part works now and which part is still simulated or carried out by a person.
Read this textbook section here
The official page below may include material before or after your assigned selection. The reader does not automatically scroll or crop to that selection.
Assigned selection — 1.6.2: Tasks: Go to 1.6.2 Tasks. Read its opening paragraphs, Figure 1.9, and four-step list. Stop immediately after that list, before the discussion distinguishing long-term knowledge and immediate belief.
Add a brief connection to your existing project update: identify the book and section, explain one relevant idea in your own words, and show the decision it informs. This is part of the same update, not another assignment.
If the official reader is unavailable or inaccessible, use the original lesson explanation and worked example, apply the same question, and label that access alternative honestly. Report the problem through the course support route; do not claim to have read an unavailable text.
Optional further reading
IDEO.org Design Kit — Build & Run Prototypes ↗
Read the five short steps. Identify what your version helps you learn and how you will record feedback. For this capstone, the final prototype must also perform the bounded task; a mock-up alone is not the final deliverable.
Try
Case file: approved Cedar Desk facts
Fictional classroom facts. Use these to check every policy claim in the exercise.
| Source | Approved fact |
|---|---|
| S1 — Session | A Standard Workshop lasts 90 minutes and costs $40 per participant, including materials. Participants must be at least 16. |
| S2 — Cancellation | At least 72 hours before the session: refund eligibility. From 24 hours up to, but not including, 72 hours: credit valid for 90 days. Under 24 hours: no automatic refund or credit; the owner reviews exceptional requests. Do not promise a payment-processing date. |
| S3 — Rescheduling | A request made at least 48 hours before the session may be considered, subject to available space and coordinator confirmation. No availability is supplied in this packet. |
| S4 — Missing facts | Ask only for the missing fact needed to apply a policy. Classroom booking references are fictional. A draft is not confirmation that a booking was changed or a refund was processed. |
| S5 — Special requests | Group prices, accessibility arrangements, age exceptions, and policy exceptions require owner review. The packet supplies no basis for promising them. |
| S6 — Release | A person checks the source, conditions, tone, and next step before a reply leaves the service. A human decides exceptions and authorizes commitments. |
Break the kit before a customer depends on it
Run three bench tests: an ordinary request to cancel 96 hours ahead; an incomplete cancellation request with no timing; and an 8-hour emergency exception. Predict the output for each before using the kit. The first should explain refund eligibility, the second should request the missing fact, and the third should prepare owner review.
| Test | Expected behavior | Observed result | Revision/retest |
|---|---|---|---|
| Ordinary | Apply S2 without claiming a refund occurred. | Record after testing. | Record after testing. |
| Incomplete | Ask for the missing time. | Record after testing. | Record after testing. |
| Exception | Hold commitment and prepare owner review. | Record after testing. | Record after testing. |
A specific revision might add an explicit “time until session” field because the first version guessed the cancellation window. Rerun the incomplete case and a normal case after the change. Correcting one path can accidentally make another harder to use.
Decision lab · practice with your Learning Companion
The demo that needs its author
Distinguish a usable artifact from a demonstration that depends on hidden help.
Take the role of usability observer. Commit to a decision, uncover evidence, revise your reasoning, then carry one action into Build. Use this as the unit’s reasoning rehearsal; reuse the response in your existing project update.
Open the AI Learning Companion section below this interactive lesson to enter this unit’s decision lab. Return here for Build. You can also use the complete reading route below.
Complete reading route · no AI service needed
Fictional teaching scenario. A fictional classmate tries your first usable version. They pause at the first instruction. Choose your next move.
- Take over and finish the task for them
- Observe, record the pause, and offer only necessary help
- Replace the task with a slide presentation
Choose one and write why before revealing the evidence.
Reveal evidence and challenge
- Observed attempt: The user cannot locate the approved source without your verbal direction.
- Repair: A visible source link and a short example input remove that obstacle in a later attempt.
- Remaining limit: The later attempt uses the same person; familiarity may also explain improvement.
Name one repair, one thing to observe in the retest, and a claim the current evidence cannot establish.
Compare with the worked reasoning
Record where the user paused and the assistance supplied. Repair the source-location step and retest the intended task, preferably with a fresh user when practical. The repaired attempt suggests a usability improvement under these conditions; it does not establish broad adoption or isolate the repair from familiarity.
Apply to your project: Write the next user task, how you will record assistance, and the specific change you will test.
Write your initial choice, evidence-based revision, and project action in your own document.
Practice records are not grades or Canvas submissions.
Build
Apply this step to your chosen problem. Cedar Desk is the supplied classroom case; keep one project throughout the course.
Open the complete classroom packet
Your continuing case: Cedar Desk
Fictional teaching case. Cedar Desk is a booking and customer-support service for small workshop businesses. Its coordinators answer questions about bookings, cancellations, rescheduling, and special requests. The operations owner wants dependable replies without unnecessary checking or repeated clarification. Your project is ReplyKit: a small source-grounded reply service that turns a permitted question into a checked response or a useful human handoff. It can run with documents, a calculator, and an available AI assistant or the supplied draft. No application or live customer account is needed.
The user is a booking coordinator. The responsible decision owner is the operations owner. A customer receives the final approved reply. Those are three different roles. The owner has agreed only to a classroom exploration in this fictional case, not to a purchase, a live integration, or a commercial launch. Your instructor may replace this packet with an approved partner brief while preserving the same tasks and evidence standards.
Approved source card: Standard Workshop policy, version 1
| Source | Approved fact |
|---|---|
| S1 — Session | A Standard Workshop lasts 90 minutes and costs $40 per participant, including materials. Participants must be at least 16. |
| S2 — Cancellation | At least 72 hours before the session: refund eligibility. From 24 hours up to, but not including, 72 hours: credit valid for 90 days. Under 24 hours: no automatic refund or credit; the owner reviews exceptional requests. Do not promise a payment-processing date. |
| S3 — Rescheduling | A request made at least 48 hours before the session may be considered, subject to available space and coordinator confirmation. No availability is supplied in this packet. |
| S4 — Missing facts | Ask only for the missing fact needed to apply a policy. Classroom booking references are fictional. A draft is not confirmation that a booking was changed or a refund was processed. |
| S5 — Special requests | Group prices, accessibility arrangements, age exceptions, and policy exceptions require owner review. The packet supplies no basis for promising them. |
| S6 — Release | A person checks the source, conditions, tone, and next step before a reply leaves the service. A human decides exceptions and authorizes commitments. |
Discovery packet
Eight supplied, fictional interview records describe waits of 4, 5, 5, 6, 6, 7, 8, and 31 minutes for a response. Six of eight describe repeated delay; three of eight say they would try a pilot. These convenience-case records are teaching inputs, not a representative market survey. They do not establish purchase intent. Response waiting time is different from the staff processing time in the later benchmark.
- Coordinator: “I often know the basic answer. Finding the latest policy and asking about exceptions takes longer.”
- Customer: “My last reply was quick, but I still did not know whether the date change was confirmed.”
- Owner: “I might try a small test if I can inspect every reply. I have not approved a budget.”
- Contrary evidence: Two of the eight records report no repeated delay. One of these customers prefers a clear policy page to another conversation.
These are invented case statements, not quotations from real people. In a discovery role-play, the respondent uses the card and says “unknown” when the packet does not answer a question. Record an actual role-play as an actual classroom event using fictional facts; do not relabel it a customer interview.
Twelve supplied test inputs
| Case | Input | Required response or handoff |
|---|---|---|
| 1 | Cancel 96 hours before the session. | Explain refund eligibility; do not claim a refund was processed. |
| 2 | Cancel 48 hours before the session. | Explain 90-day credit eligibility, not a refund. |
| 3 | Cancel 8 hours before the session because of an emergency. | No automatic promise; send the exception to the owner. |
| 4 | Cancel, but the time until the session is missing. | Ask for the missing time before choosing a policy branch. |
| 5 | Move a booking 60 hours before the session. | State the space/confirmation conditions; do not confirm a seat. |
| 6 | Move a booking 20 hours before the session. | Escalate rather than extend the published rescheduling rule. |
| 7 | Book for a 14-year-old. | State the age condition; an exception needs owner review. |
| 8 | Ask the duration and whether materials cost extra. | Give 90 minutes and $40 including materials. |
| 9 | Cancel exactly 72 hours before the session. | Apply the refund-eligibility boundary correctly. |
| 10 | Cancel exactly 24 hours before the session. | Apply the credit boundary correctly. |
| 11 | Ask whether a particular access arrangement is guaranteed. | Ask the owner to confirm; do not invent an arrangement. |
| 12 | Request a special price for a group of 20. | Prepare a quotation request for the owner, not a made-up price. |
A complete usable example
Intake: Case 5; request to move a booking 60 hours before the session. Sources: S3, S4, S6. Draft: “Your request is at least 48 hours before the session, so a change can be considered. It depends on available space and coordinator confirmation. Which session would you like us to check? Your booking has not yet been changed.” Review: Timing, conditional wording, missing desired session, and no false confirmation checked. Handoff: A coordinator checks availability and confirms or declines the requested change. Status: Approved draft for classroom use; no external message sent.
Make the example usable by another person: provide this source card, an intake sheet, a draft-and-source table, the review checklist, an exception note, and a delivery record. Number the version and label who performs each step. A slide showing those items without providing them is not a working kit.
Keep four datasets separate
| Dataset | Values | Evidence limit |
|---|---|---|
| Discovery | Eight waits; six report repeated delay; three would try a pilot. | Supports questions about this fictional sample, not demand in a population. |
| Processing benchmark | Same 12 cases: manual 96 minutes and 12 acceptable replies; unchecked AI 24 and 9; reviewed AI 84 and 12. | Supplied performance comparison, not your actual user test. |
| Monthly planning scenario | 1,200 eligible replies; 50% planned adoption; one minute released per assisted reply; $24/hour capacity value; $180 tools plus $60 support monthly. | A conditional forecast. Adoption, transfer of timing, and costs require verification. |
| Adoption cohort | 12 invited, 9 started, 6 completed, 4 returned within 14 days after invitation. | Return means completing another task in that same 14-day window; not a forecast for all coordinators. |
Your actual evidence starts with what you do. Use the packet to practice and build. Then ask a classmate, instructor, or approved partner user to perform a bounded task with your version. Record the real attempt, assistance, result, revision, and retest. Leave fields blank until the event happens. A self-review is useful preparation but is not another person’s user test.
Prepare Milestone 2: the first usable solution
Give the file to a classmate or approved reviewer and ask them to locate the policy, complete one fictional intake, and produce the required output. Observe where instructions fail. Record any explanation you had to provide. This brief access/use check prepares the more deliberate comparison in Units 6–7.
Save the version, bench-test record, meaningful revision, and retest. Include the actual kit, not only screenshots. If the reviewer cannot complete the bounded task, revise the instructions or scope. If no reviewer is available yet, mark the user check pending and use the arranged classroom-review route in the course guide. In a facilitated offering, request a reviewer by Unit 4 and confirm the pilot and retest by Unit 6. A timely request delayed by course staffing receives an adjusted evidence deadline rather than a penalty for unavailable review. An author’s self-test is not an independent attempt.
Submit or review the same project snapshot using Milestone 2’s criteria. A document-operated AI-assisted service or clearly labeled supplied-output simulation can meet the standard when its input, AI contribution, checked output, and handoff can be inspected. Its manual fallback supports continuity; it is not evidence of live AI integration. A mockup that cannot produce or hand off an output cannot. Explain where AI assisted and where a human corrected it. Next, design a fair test that compares complete effort and quality rather than impressions.
Add your project evidence and reflection to the check-in in Canvas.
Your project record
Write here or use your own document. Notes stay in this browser; export a backup to keep them. Nothing is submitted automatically.
Finish
What will you carry forward?
Your Canvas check-in
Your attractive prototype screen has no usable output, source card, or manual process behind it. Which addition most directly meets the implementation goal?
Use the evidence from this lesson to explain your reasoning in Canvas. Continue with Canvas Next below; submit your check-in in Canvas.
The simulation and Learning Companion support practice. They do not submit work, record Canvas completion, or assign a grade.
In your project record, state what you decided, the evidence you used, and one uncertainty to test next. Explain the reasoning in your own words.
Milestone 2 of 5
First usable solution
Save a snapshot of your project record. In a taught offering, submit it using your instructor’s arrangements and apply the feedback before the next dependent task.
Evidence and assessment criteria
Provide the actual versioned artifact or supervised service kit, its chosen AI capability and operating workflow, three varied bench tests, a brief reviewer access/use check, and a meaningful revision with retest. A document-based service is sufficient when another person can complete the bounded task. A mock-up or demonstration deck alone is not.
Include the source card, input/output examples, review criteria, exception handoff, and manual fallback. Identify the permitted input, AI contribution, checked output, and implementation route. A supplied-output route must label its AI step as simulated; do not claim a live model connection or automation. Record assistance honestly. If another person has not attempted the task, show that requirement as pending rather than substituting an invented result.
| Criterion | Maximum points | Meets the goal | Strong transfer | Scoring anchors |
|---|---|---|---|---|
| Usable end-to-end task | 35 | A reviewer accesses the actual version and completes the bounded task; any assistance and unresolved limitation are recorded. | An unfamiliar reviewer completes a changed input using clear instructions, and the author explains why the design transfers. | Strong evidence: 35; Meets the goal: 28; Developing: 17.5; Not demonstrated: 0. Intermediate points within this criterion's maximum are allowed when the instructor explains the criterion-specific evidence. |
| Grounded AI use and human control | 25 | The chosen capability fits the task; permitted inputs, the AI output, live or simulated execution, reviewer, source boundary, exception, and fallback are explicit and checkable. | Shows a consequential AI error or limitation and an effective correction without relying on the assistant’s self-confidence. | Strong evidence: 25; Meets the goal: 20; Developing: 12.5; Not demonstrated: 0. Intermediate points within this criterion's maximum are allowed when the instructor explains the criterion-specific evidence. |
| Testing, revision, and retest | 25 | Ordinary, incomplete, and exception cases have expected and actual results; a meaningful defect is revised and retested. | Tests whether the change affects another path and explains residual uncertainty or an alternative cause of failure. | Strong evidence: 25; Meets the goal: 20; Developing: 12.5; Not demonstrated: 0. Intermediate points within this criterion's maximum are allowed when the instructor explains the criterion-specific evidence. |
| Operating clarity and evidence record | 15 | Version, sources, steps, AI capability and contribution, live or simulated execution route, observations, and pending actions are understandable and accurately labeled. | A concise change log lets a second person reconstruct what changed, why, and which evidence supports the current version. | Strong evidence: 15; Meets the goal: 12; Developing: 7.5; Not demonstrated: 0. Intermediate points within this criterion's maximum are allowed when the instructor explains the criterion-specific evidence. |
How criterion scores are assigned
Each rubric criterion keeps its published maximum points. The instructor judges the evidence against that criterion's Meets the goal and Stronger evidence descriptions and uses these numerical anchors.
| Evidence level | Anchor | What the evidence shows |
|---|---|---|
| Strong evidence | 100% of the criterion maximum | Meets the goal and demonstrates the criterion's stronger evidence or transfer. |
| Meets the goal | 80% of the criterion maximum | Demonstrates the stated goal with accurate, usable evidence. |
| Developing | 50% of the criterion maximum | Shows a relevant attempt, but important evidence is incomplete, unreliable, or needs correction. |
| Not demonstrated | 0% of the criterion maximum | Provides no usable evidence for this criterion. |
These are judgment anchors, not automatic all-or-nothing ratings. Intermediate points are allowed within the criterion maximum, with a comment identifying the evidence and remaining gap. For a 25-point criterion the anchors are 25, 20, 12.5 and 0 points. A defensible decision to narrow, revise, or stop can earn strong credit. These anchors do not change criterion maximums, assessment totals, course weights, or previously recorded scores.
A favorable project result is not required; the quality of evidence and judgment matters.
Before saving this snapshot
You can revisit any step. Completion records study progress; it does not confirm a submission, grade, or project quality.