POLISHING UI/UX
Want to back the stream? The «داعم» membership on YouTube is EGP 60 a month: badges, emoji, and I answer your questions first on stream. Become a supporter
The gist
The stream follows a Shopify app moving from its core build into the harder pre-launch phase of polish and verification: UI/UX review, turning findings and screenshots into tasks, and manually exercising the app alongside AI agents. Luqman presents a command center that divides work among the human operator and agents, balancing parallel execution against file conflicts, and proposes teaching a lighter version from scratch. Practical investigation of slow loading and navigation occupies a major part of the work; routing, repeated database calls, Prisma, and network distance are considered, and some improvement is observed, but the complete cause and final fix remain unresolved. App-listing requirements, privacy policy placement, GDPR, Railway deployment, and the path toward Shopify submission are also discussed as unfinished work or plans rather than confirmed outcomes. Other useful discussions cover competitor research and positioning in a low-competition market, subjective tradeoffs among Codex, Claude, and other models, usage limits and subscription ROI, plus audience questions about Arabic dictation, email outreach, and channel growth through short-form video. The stream also includes attributed views on AI and cybersecurity and on how operator expertise shapes the value of AI tools.
73 moments from the stream
- 0:00:42
Building a Command Center from Scratch
Luqman proposed a practical walkthrough that builds a command center from scratch, following an earlier explanation of what it is and why it is useful. He said the setup may be simpler than people expect.
- 0:01:45
Application Polish and UI/UX Review
The application was described as nearly complete, with an overall review, UI/UX polish, and debugging still required. The plan was to use the app directly and collaborate with AI agents to find issues, while acknowledging that debugging can be tedious.
- 0:03:59
Preparing an App for Shopify Submission
Responding to a progress question, Luqman said the core build was nearly finished and that the remaining work included polish, debugging, review, and completing Shopify’s pre-submission checklist. As a contemporaneous estimate, he said acceptance review might take roughly two to four weeks.
- 0:05:33
Dividing Work Between a Human and AI Agents
He explained a visual command-center system that distinguishes operator-owned, agent-owned, and collaborative tasks, noting that the pre-submission milestone includes substantial hands-on work that should not all be delegated.
- 0:06:06
Navigating Shopify Listing Requirements
He said the application still needed polish, review, and debugging, and raised the question of taking screenshots, though that thought is incomplete in the captions. He then requested step-by-step guidance in the Shopify dashboard for locating listing requirements and mentioned the privacy policy; some field names and surrounding phrases are unclear.
- 0:08:32
Proposed Workflow After App Submission
He outlined a plan contingent on finishing the remaining work: submit the app to Shopify, wait for the review outcome, then return for further polish, marketing, and improvements to the app’s website. This was a future plan, not a completed result.
- 0:10:20
An Interactive Workflow for UI Review
Explains that reviewing the interface and trying the app requires more back-and-forth with agents than the usual linear task workflow because some steps must be performed and inspected manually.
- 0:12:02
Limits of AI-Guided Website Navigation
Discusses having an agent follow browser steps while noting that model guidance can become confusing or inaccurate as website interfaces change, including dashboards such as Shopify and advertising platforms.
- 0:15:49
Finding Where to Add a Privacy Policy URL
Discusses guidance indicating that a privacy-policy URL is added through the app listing and submission flow, with a suggestion to prepare and host the policy content in advance; the location is not actually verified on this page.
- 0:17:49
Turning UI/UX Findings into Structured Tasks
Proposes trying the app from customer and merchant perspectives, then using a milestone node to capture each discovered issue as a labeled task before analyzing the debugging process and reusing the workflow for fixes.
- 0:26:48
Troubleshooting a Failed App Registration Transaction
After an app-registration transaction repeatedly failed without a clear explanation, IP or regional restrictions were considered as an unconfirmed possibility. Testing a VPN and completing registration were deferred until off-stream to avoid disrupting the stream.
- 0:29:41
Working Around Blocked Tasks and Continuing Development-Store Setup
The workflow was to set aside tasks requiring human intervention, request that the remaining work be unblocked, and continue with a task involving development-store setup and app installation. The speaker believed part of that setup might already have been completed.
- 0:31:56
Organizing Parallel Tasks with Codex
The speaker demonstrates a workflow for preparing several tasks and running them in parallel with Codex while adjusting the reasoning level, describing it as a reusable pattern for startups and SaaS products. He also raises a question about why some tasks are blocked and the consequences of unblocking them, but this page provides no definitive answer.
- 0:35:58
Including GDPR Requirements in Application Planning
The speaker mentions a task for a GDPR notice aimed at users or merchants operating the application in Europe, noting that the tool inferred this requirement from the project manifest and roadmap even though he had not planned it himself.
- 0:35:58
Beyond the Initial Scaffold: The Human Review Role
The speaker explains that AI can propose an application's base scaffold, roadmap, and previously unconsidered requirements, while emphasizing that the subsequent human work is to inspect the output, review it, and direct the tool toward the needed improvements.
- 0:36:57
Sequencing Product Polish Before Shopify Review
The speaker outlines a proposed sequence: polish the UI first, improve the UX next, and then continue with the backend, because he does not want to build on an unclean interface. After the remaining tasks, permission review and the Shopify app review and submission flow are expected to follow.
- 0:39:54
Reviewing Codex Usage Across Multiple Tasks
The speaker plans to review Codex usage after three parallel tasks finish in order to assess the relative consumption of this workflow. The page ends before any measurement or conclusion is presented.
- 0:40:24
Choosing AI Subscriptions by ROI and Usage Limits
The speaker discusses how usage limits can interrupt intensive workflows and argues that additional tool or API subscriptions should be evaluated as investments tied to workload and expected return, rather than costs incurred without a clear plan.
- 0:44:59
How Parallel Agents Avoid File Conflicts
While monitoring parallel tasks, the speaker explains that agents can see one another’s work state and flag an active dependency or the risk of editing the same file. Progress and small milestones are recorded in a JSON file, helping them detect overlap and choose to wait for the dependent task before proceeding.
- 0:48:05
From App Completion to Shopify Deployment and Debugging
On April 9, the speaker says the application is nearing completion earlier than he had expected, with polishing work remaining. He outlines the next steps as finishing the active tasks, deploying the update to Shopify, launching the app, beginning debugging, and reviewing coding-tool usage after the task completes.
Got a question about something in this stream?
Ask in the group