FIXED 1 BUG. CREATED 3 MORE.
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
A wide-ranging working stream focused on reviewing a Shopify app before launch, combining chat and merchant-UX refinement, fault diagnosis, and conversion of review findings into a debugging and testing queue inside a custom command center. Practical work covered crashes caused by products without images, disappearing text or attachments, slow replies, alignment problems, contextual actions, and confirmation states. A recurring conclusion was that a local fix is not proof of system readiness: human review and realistic scenario testing remain necessary. Luqman also explored agent-driven debugging through backend tests and log inspection, context management through documentation, state files, and fresh sessions, plus confirmation gates and Git checkpoints to constrain autonomous changes. Audience questions led to substantial discussion of Claude, Codex, OpenCode, OpenRouter, local models, and agentic coding tools. The practical view was that model choice depends on the task, quality, cost, and familiarity with the project, while benchmark tables must be interpreted through their methods rather than a single ranking. Many multi-model and cheaper-routing ideas remained proposed experiments rather than completed findings. The stream also covered validating product ideas with landing pages, taking products to market instead of repeatedly building, and moving from a useful first release toward scaling based on real usage. Visual brainstorming and interface design, SEO and GEO, precise product descriptions, and an Arab knowledge-sharing community were additional useful threads. Open questions remained around some latency and timeout causes, classifier-context cost trade-offs, automation reliability, and several fixes and tool trials that were not conclusively validated.
155 moments from the stream
- 5:22:53
Investigating Broken Chat UI and Logic
The speaker identifies several chat failures: messages aligned on the wrong side, text that does not wrap, and an empty reply followed by two unexplained buttons. He proposes a full investigation and dividing the chat-box UI inspection among multiple subagents.
- 5:23:50
Trying Alternative Models for Frontend Work
The speaker discusses the learning curve of switching models after becoming comfortable with Claude and Codex, while viewing alternatives as useful for gaining experience or handling smaller tasks. He proposes trying GLM to polish the relatively empty “Open the Assistant” section, qualifying that its design reputation is something he has heard rather than a result demonstrated in this excerpt.
- 5:32:59
Refining a Status Card in the UI
A UI card review favors contrast with the neighboring dark card, preserves the Polaris card structure, and considers an animated status indicator.
- 5:36:26
Deployment Workflow with GitHub and Railway
An explanation of a workflow where only significant changes are pushed to GitHub, after which the new commit automatically deploys the app to Railway.
- 5:38:33
When Productivity Tooling Becomes Overbuilt
The discussion considers automating branching, merging, and pushing from a command center, then explains why integrations and polish that do not materially improve efficiency were dropped. The goal is to keep the tool focused instead of gradually turning it into a full IDE.
- 5:40:12
The Failure Was Deeper Than the UI
After strange problems appeared while several areas were being changed together, the proposed check was to reproduce the interaction directly in the app. The chat crashing and no longer responding indicated a root cause deeper than visible UI breakage.
- 5:45:17
Why a Quick Fix Can Create New Bugs
The speaker explains that debugging can uncover or introduce further problems, especially when a model is asked to fix an isolated area without enough context. Side effects must therefore be checked across the system rather than treating a local change as the end of the work.
- 5:46:05
Testing Real-World Use Cases Before a Public Launch
A public application will face varied usage scenarios that are difficult to anticipate in full, so the work focuses on checking multiple parts of the product before a first release to the Shopify App Store. The speaker also gives the opinion that genuine value and product quality should take priority over immediate concern with marketing and adoption.
- 5:47:20
Clear Value as the Basis for Selling Credits
The speaker argues that users will buy and use credits when they receive substantial value, provided the product is polished and good.
- 5:50:52
Disappearing Screenshot and Launch Reliability
The speaker describes a bug where an attached screenshot appears briefly and then disappears. He discusses the need for a fallback strategy and strong reliability when launching a Shopify app, especially for early users, while separate sessions work on response speed and the bug; the page does not confirm that the fix was completed.
- 5:54:51
Consistency and Ambition in Daily Work
In an audience discussion, the speaker says patience and consistency create cumulative progress even when daily steps are small, and links sustained effort with ambition and a clear vision. This is presented as his opinion, not as a demonstrated test result.
- 5:58:59
Why review and debugging can reveal new problems
The speaker distinguishes verification and debugging tasks from a separate list of actual issues. Based on this work session, he observes that review or debugging can expose additional problems, making the process iterative rather than a one-pass exercise.
- 6:02:08
Exploring an assistant section design with Superpowers
The speaker tries the Superpowers plugin to start a brainstorming session for redesigning an assistant prompt-card section. The tool first asks whether the section should prioritize discovery or quick launch; he says both matter but leans more strongly toward quick launch.
- 6:07:22
Inserting a Person into Thumbnail Designs and Refining the Result
A practical review of a tool that inserts a user’s photo into compositions inspired by successful thumbnails. The discussion covers supplying a higher-quality image, refining artifacts with prompts such as fixing a hand, and avoiding overly close copies of recognizable thumbnails that could raise copyright concerns. It also includes advice to begin marketing completed applications rather than only building them.
- 6:18:51
Visual Brainstorming Before UI Implementation
The speaker demonstrates using Superpowers Brainstorm to produce a visual concept—potentially an animated SVG—and preview it through a local port before implementing the UI close to the mockup. He also specifies a flat list that could later be sorted by popularity.
- 6:22:25
Practical Feedback for an MVP
The feedback praises the design but recommends adding substantially more templates and covering more niches, while acknowledging that the product is still an MVP that can expand over time.
- 6:27:08
Don’t Overbuild: Ship Version One
A concise recommendation to avoid overbuilding, complete a version one, and deploy it to the market instead of delaying launch.
- 6:27:38
Tracing Disappearing Text in a Chat UI
The speaker tests a simple phrase after observing that text appears briefly and then disappears, checking the hypothesis and response timing before settling on the bug’s cause.
- 6:30:18
Why the Proposal Does Not Feel Clickable
The speaker rejects the proposal because it feels too generic and does not seem distinctive or compelling enough to click.
- 6:30:58
App-Store Release and the Start of Marketing
The discussion treats publishing an app to an app store as a practical milestone and warns that overbuilding can stall progress. It also encourages developers who have shipped apps to exchange experience, especially around marketing and store-submission requirements.
Got a question about something in this stream?
Ask in the group