System Prompt Downstream Funnel
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 combining a Burnish review and submission readiness with discussion of AI-assisted product building. Practical work covered installation, scopes, billing, the storefront, tiered human-approved Autopilot optimization, the brand manifesto, traffic and revenue dashboards, pixel/GA4 integrations, and meaningful empty states. Some narrow checks worked—such as a level transition and selected product-data updates—but the stuck optimization/publishing flow and missing analytics remained unresolved, prompting discussion of tunnel vision, error monitoring, debugging orchestration, and subagent workflows. The stream also explored product-scope discipline, capped trials, market and competitor validation, unit economics, and learning software development while building with AI. Luqman argued that deep software-lifecycle expertise is not an absolute prerequisite for starting, while making clear that AI tools do not guarantee bug-free production software. Audience questions additionally touched on APIs versus webhooks, local and future AI models, and context-window limits, alongside lighter conversations about stream music, World Cup predictions, and a virtual-office idea for tracking project progress.
69 moments from the stream
- 0:01:07
Prioritizing Core Work and Preventing Scope Creep
A reflection on prioritization mistakes, favoring UX and frontend–backend integration over continually adding features. The central lesson is to set a stopping point for feature work so the product can actually be submitted.
- 0:03:41
Visual QA Before App Submission
An explanation of submission readiness through a visual walkthrough of the app, inspection of its pages and sections, and resolution of remaining defects, with attention to the areas delivering the core user value.
- 0:05:24
Coding Sessions That Spawn Follow-Up Work
A coding-agent workflow in which a session can start a follow-up session after completing its task when it identifies additional work or requirements that should be handled before merging.
- 0:08:11
Testing Synchronization with Mock Data and Monitoring Real-World Errors
A development store and generated data are used to test product behavior, including whether deletions propagate automatically. The discussion notes that real-user testing exposes edge cases that are difficult to anticipate completely, while error monitoring can capture failures when they occur.
- 0:10:22
Adding New Products and Resetting the App
He outlines a proposed workflow: temporarily tag existing products, add new unoptimized products, remove the tagged products, and then reset the app. He says the storefront section produced many issues because it involved substantial backend wiring and editing, while expecting the remaining pages to be simpler; the final outcome was not yet confirmed.
- 0:13:29
A Gamified Virtual Office for Tracking Project Progress
He presents a side-project idea for helping active coding-community members follow one another’s project progress. He imagines starting with a small virtual office, adding gamified elements, expanding it later if needed, and possibly sharing the repository with people interested in improving it.
- 0:19:10
Where Is a Connection-String Password Configured?
A technical question asks where the password referenced by a connection-string URL should be configured. This page does not capture a confirmed answer, and no credential value is exposed.
- 0:22:05
Showing Progress and Automating a Shorts Workflow
A practical note on presenting work progress more clearly and a desire to automate the short-form video workflow. One task was said to consume many tokens, though the speaker felt it might be worthwhile.
- 0:25:12
Resetting an App and Reviewing the Storefront Flow
A work plan to remove the current items, reset the app, reinstall a branch, and review scopes, onboarding, the storefront process, and its workflow before completing it. A suspected issue in the storefront process remained uncertain, with no confirmed resolution.
- 0:27:06
Tunnel Vision When Fixing Bugs
An explanation of the risk of focusing too narrowly on one bug: a change may remove the visible problem while silently creating others in the backend. The speaker considers this especially difficult without a programming background. The speaker expressed concern that changes made across multiple coding sessions might increase bugs rather than reduce them.
- 0:28:20
Do You Need Software-Lifecycle Knowledge to Build with AI?
Answering an audience question, the speaker argued that detailed software-lifecycle knowledge is not strictly required; learning how to use and communicate with AI—including agentic workflows and protocols—is more important. He noted that working with AI has a learning curve and varies by person, and that a programming background leads to a different way of working with AI, then asked what the questioner specifically meant by the software lifecycle.
- 0:31:11
Learning Software Development by Building with AI
A discussion of how AI tools have lowered the barrier to building applications. A beginner need not master frontend, backend, and databases beforehand, but can gradually learn concepts such as GitHub workflows, APIs, testing, branches, and worktrees while building a real project.
- 0:33:41
Building to Learn vs. Validating a Business
The speaker distinguishes building personal tools for learning—where almost any useful workflow idea can be a valid starting point—from building a product to sell, which calls for research into the market, profit margins, unit economics, and product potential before implementation.
- 0:36:57
Validate the Problem, Demand, and Competition Before Building
An explanation of idea validation: identify the problem and value proposition, then examine market activity and competitors before building. In the speaker’s view, having no competitors is not always positive; it may indicate a hidden obstacle or weak demand, potentially requiring product education and demand creation.
- 0:39:46
Blue Ocean and Red Ocean Markets
A brief, incomplete introduction to blue-ocean markets, where competition is absent or limited, and red-ocean markets, where the field is crowded with competitors.
- 0:40:21
Choosing a “Purple Ocean” and Competing Through Execution
The speaker describes a “purple ocean” as a middle ground between a new market and a crowded one, suggesting it can be a useful opportunity. An existing idea is not necessarily a bad choice: a stronger execution may compete in a market whose demand is already established.
- 0:45:06
Readiness-Score Consistency After Product Deletion
During a live check, the speaker observed that the product count changed after deletion while the readiness score did not. He framed this as an engineering question about correctly recomputing the score, especially after removal of a manual recompute action; the page does not show a confirmed resolution.
- 0:47:21
Product Optimization and Change-Review Workflow
The speaker explains a workflow in which a merchant selects products for optimization, reviews the generated changes, and can regenerate them before publishing them to the products. Optimization and publishing can also be applied to a selected group, and he says competitors use a similar flow.
- 0:49:04
Progressive Autopilot With Human Approval
The speaker describes an autopilot mode with progressively increasing trust. At the first level, a batch of products is selected and optimized without automatic publication; the merchant must review and publish the results before advancing. Details of later levels are not completed on this page.
- 0:50:22
Trust Levels Before Full Automated Publishing
An explanation of graduated automation that begins with a mix of automatic publishing and human review, then expands autonomy after a review history. The aim is to let a merchant establish confidence in output quality before enabling automatic operation, with a log of recent tasks and activity.
Got a question about something in this stream?
Ask in the group