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
- 3:00:41
Speed Debugging Expands into Many Tasks
After reopening the work interface, the speaker observes that investigating a speed-related problem produced many tasks, suggesting that the performance-debugging scope is larger than it first appeared.
- 3:02:42
Grouping Independent Tasks for Parallel Work
He reviewed the work history and what had been completed, then identified groups of independent tasks that could be prepared and run in parallel across multiple terminal sessions.
- 3:07:34
Step-by-Step Railway and Supabase Alignment Check
He requested an interactive, one-step-at-a-time process for checking Railway and Supabase alignment, beginning with the Railway region instead of receiving a long list of instructions at once.
- 3:08:19
Choosing an AI Model by Workload
He offered a subjective view that model choice depends on the task, comparing Claude, GPT, and Gemini for coding, pragmatic work, and translating PDFs into Egyptian Arabic, while mentioning GLM-5 as a strong option for sufficiently capable hardware. No benchmark testing was presented.
- 3:09:35
Checking the Railway Region During Troubleshooting
He opened Railway's settings and checked the selected region, then tentatively concluded that the issue did not appear to come from that region setting; this segment did not establish an alternative cause.
- 3:13:51
Why a Debugging-Agent Review Is Not Enough
The speaker argues that an initial build may be relatively quick, while making the application correct and stable can take much longer. Code and security reviews can be divided among debugging-agent sessions, but the orchestrator still needs to use the app personally to uncover UI, UX, and backend problems across its different paths.
- 3:18:13
Checking Deployed Sections Reveals Slow Navigation
After deploying the app and restarting the server to inspect newly added sections, the speaker observes that responsiveness is worse than before and that the delay obstructs navigation, shifting the work back toward speed investigation.
- 3:24:07
A Pre-Launch Application Quality Audit Service
The speaker examines an external service where a product can be submitted before launch for an application audit and quality assessment. He considers it potentially useful for other app projects, while the service name and some details remain unclear in the captions.
- 3:25:56
Diagnosing Slow Navigation with Instrumentation and Deeper Research
The severe slowdown is investigated collaboratively with an agent, including a temporary error followed by the app working again. The agent’s analysis suggests Prisma queries or database connection handling as possible contributors, but the excerpt does not establish the cause, so the speaker requests instrumentation and deeper research in pursuit of near-instant navigation.
- 3:32:17
AI Broadens Access to Software Building
The speaker argues that AI tools are opening software development to people without a programming background, helping them turn market knowledge into applications they can build.
- 3:35:26
Diagnosing Development Latency from a Remote Database
The speaker considers whether the app feels slow because the development machine in Egypt is communicating with a database hosted in a US region. He asks how to verify that hypothesis and temporarily improve local development, with running PostgreSQL locally appearing as a proposed step; the excerpt does not establish a confirmed result.
- 3:42:24
Moving Toward a More Interactive LLM Workflow
The speaker reflected that each stream produces some progress even when there is no obvious major result. He expected upcoming work to involve more back-and-forth with language models and a less straightforward workflow, including practical checks of whether an agent could perform tasks such as deployment.
- 3:44:21
Testing Deployment of an Updated App Through Railway
After many updates during the week, the speaker tested whether the agent could handle deployment instead of doing it manually. The discussion covered a large set of uncommitted changes, deciding what should be committed, and pushing so Railway could pick up the update, while leaving the outcome uncertain and noting that deployment might take some time.
Got a question about something in this stream?
Ask in the group