Day 5 Summary Codex AI Architect Technical Design

Day 5. Service Design (UI/UX Review) Course Summary

Master an architectural UI/UX design approach: find usability weak points from the user's view based on screen specs and path design, review layout consistency between screens, and shorten the flow depth to reach the final goal.

Part 1. Background
Part 2. Lab goals
Part 3. Follow along
Part 4. How it works
Part 5. Teaching guide
Part 6. Final Task

Barriers in usability design

In real projects, misunderstandings between planning, design and development lead directly to missing database items and broken consistency. Let's see why guaranteeing consistency at the design stage matters.

Understanding gap

Stubborn feature-centered design means first-time beginners don't understand the UI and quit the app midway.

Vocabulary gap

Button positions, sizes and labels (Username vs Customer ID) differ on each screen, making users tired of relearning.

Sharing gap

Users go through an overly long 9-step screen journey to reach their goal (booking a session), lowering final payment conversion.

ChatGPT vs Codex Client (Workspace-based)

Compare ChatGPT, limited to a single chat context, with Codex Client, which understands the whole folder structure and dependencies between files.

Compared ChatGPT (general chat AI) Codex Client (Workspace-based AI)
Context awareness Compares and analyzes many screen plan files and the path depth of user entry together Traces mismatched UI button designs, sizes and alignment, and status message formats across login, sign-up, My Page and more
Screen consistency review Gives only convenience feedback on a single text screen Recommends consistent standard names and DB column names based on the project term dictionary
Impact analysis Suggests isolated schema errors or made-up table fixes Works backward from small schema changes (e.g. adding a column) to list API specs and screen parts to fix

Summary of today's final outputs

The list of final outputs learners create during the Day 5 session, and the purpose of each.

Task 01. Data structuring

A Mermaid ERD script and spec rendering entities and relationships pulled from a natural-language plan (output: erd-schema.md)

Task 02. Screen consistency review

A project standard term dictionary defining scattered planning terms as standard English columns (output: project-glossary.md)

Task 03. User journey improvement

A change impact analysis listing API and screen fixes for adding booking approval (WAIT_APPROVE) (output: change-impact-report.md)

Mission 01. Design record

A Design Decision Log (DDL) archiving the reasons for normalization and data design decisions and rejected alternatives (output: design-decision-log.md)

Mission 02. Evaluate service quality with AI

An Edge Case analysis report with DB improvements for concurrent payment errors, data lost on account deletion (referential integrity) and transaction rollback (output: edge-case-analysis.md)

Input / Output data pipeline

A mapping of the inputs used in the lab and the outputs produced through Codex Workspace reasoning.

01
Data structure
02
Terms
03
Change impact
M1
Decisions
M2
Exceptions
Task 01. Review screens from the user's view
šŸ“„ Input (planning meeting notes)
• day5_meeting_notes.md (meeting notes) - User flow scenarios and booking process as unstructured prose
Codex Workspace āž”
šŸ’¾ Output (ERD spec)
erd-schema.md with Mermaid ERD code - Visualizes Primary Key/Foreign Key relationships of the User, Expert, Reservation and Payment entities

STEP 1. Task 01 - Review screens from the user's view

Enter the meeting scenario prose and derive precise normalized relationships (1:N, 1:1) and Mermaid code.

šŸ” Codex request prompt (data structuring)

Based on the screen spec of the provided booking request form, review whether the information structure is clear from the view of a beginner who's new to the service, point out what hurts usability, and write an improvement report (UX Review Report).
  • Check that the PK and FK settings for the 4 entities User, Expert, Reservation and Payment don't conflict with the scenario logic.
  • Check that cardinality notation is correct when the Mermaid diagram (`erDiagram`) renders.

STEP 2. Task 02 - Screen consistency review

Compare the gap between Figma plan terms and DB dev specs and build a domain term master dictionary.

šŸ” Codex request prompt (term dictionary)

Compare the screen designs in the provided day5_signup_flow.md with each other, analyze elements that lack consistency in button placement rules, terms used and feedback messages, and write a screen comparison report.
  • Check that standard English names and column types that fit the business domain are given correctly.

STEP 3. Task 03 - User journey improvement

Find the API and table impact when the release plan changes to add an 'approval' step to booking.

šŸ” Codex request prompt (find impact)

Analyze the detailed booking scenario path in the provided day5_signup_flow.md and write a User Journey Report that minimizes the click steps and screen changes a user needs to complete a session booking.
  • Check that it detected the chain impact of the Reservations table status change, new API specs and button state transitions on the planned screens.

STEP 4. Mission 01 - Be a first-time user

Log the rejection reasons and agreements for normalization decisions, splitting booking history physically, and so on.

šŸ” Codex request prompt (decision record)

You are a self-employed person in your 50s who is a bit clumsy with smartphone apps. Looking at the sign-up and first screen UI spec in the provided day5_signup_flow.md, write from your own point of view what is most confusing and hard when signing up, with your emotional reactions and requests for improvement.
  • Check that the reason for splitting the history table (DEC-003) and the alternative (slower performance with one table) are clearly logged.

STEP 5. Mission 02 - Evaluate service quality with AI

Audit system weak points to prevent referential integrity violations (account deletion) and broken transactions.

šŸ” Codex request prompt (derive Edge Cases)

Based on the provided day5_signup_flow.md spec, score each of 1) information readability, 2) control accessibility, 3) error recovery and 4) flow efficiency out of 5, and build a service quality evaluation (Usability Evaluation) with an overall score, strengths and painful areas to improve.
  • Check the plan to keep referential consistency with Soft Delete (logical delete) and the response to distributed transaction rollback.

How It Works & Key Mechanisms

Schema-aware reasoning

Going beyond simple text matching, it understands entity referential integrity (FK) and transaction lifecycles, and works backward from a natural-language plan to DDL (Data Definition Language) queries and state machine rules automatically.

Meta term clustering

It groups differently written terms (customer name vs Nickname) by meaning (Clustering) to produce a company-wide term master dictionary, and syncs the data dictionary's consistency based on it.

How the five Tasks/Missions work together in technical design

See how data and analysis from each step go around the technical design loop to maximize quality.

[Beginner usability review] (Task 01. Find pain points)
ā–¼
[UI style consistency check] (Task 02. Shared rules)
ā–¼
[Shorter User Journey design] (Task 03. Flow optimization)
ā–¼
[Collect virtual customer struggle logs] (Mission 01. First-impression feedback)
ā–¼
[Usability score diagnosis] (Mission 02. Score and fix)

This usability diagnosis pipeline removes usability errors and design contradictions before a single line of code is written, making sure users can understand and use the product easily.

šŸ’” Course design specs

  • Difficulty: ā˜…ā˜…ā˜…ā˜…ā˜† (advanced - includes database normalization and distributed transaction concepts)
  • Lab time: 1h 30min ~ 2h (combined design lab)
  • Real-world use: ā˜…ā˜…ā˜…ā˜…ā˜…

Learner questions/discussion topics

Topics that work well for class discussion or interview questions.

Q1. Account deletion strategy

"When a sign-up error wipes out all entered data, how should we balance the security view (deleting data) and the user experience (UX) view?"

Q2. Terminology Debt

"If a project runs to the middle without a screen consistency guide, how does the resulting 'terminology debt' inside the dev team affect the real schedule?"

Final course Task for the planning/design stage

An overview of all the lab tasks and missions covered on Day 5.

Task Topic How AI is used
Task 01 Review screens from the user's view Have AI review awkward screen layouts from a beginner's view and suggest fixes
Task 02 Screen consistency review Compare term use and button position symmetry across pages and set rules
Task 03 User journey improvement Design a flow that cuts unnecessary standalone screens to reduce user clicks
Mission 01 Be a first-time user Use an older persona unfamiliar with smart devices to recreate sign-up struggles
Mission 02 Evaluate service quality with AI Evaluate usability, understanding and convenience together and convert them to grades and scores
← Enter Day 5 Classroom View full roadmap