AI COMMERCE INDEX SCORING ENGINE
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 documents a build-in-public effort around an engine for measuring how visibly Shopify merchants’ products appear in AI answers, alongside a Command Center for classifying agent tasks by size and dependency, executing them, and reviewing them through task- and milestone-level audits. Practical work covered code-informed prompt preparation, correcting illogical dependencies, monitoring tasks and logs, and improving the README and installation options. Although some workflows were described as becoming more autonomous, defect review, the frontend, and several implementation decisions remained unfinished. Luqman also compared Codex, Claude, and other tools from personal experience, consistently qualifying observations about speed, quality, resource use, and session variability as non-benchmark impressions rather than settled conclusions. Audience questions and wider discussions covered market validation and marketing before Shopify approval, organic outreach versus paid acquisition and waitlists, landing-page structure, animated backgrounds and calls to action, and GEO/AEO compared with traditional SEO. Other useful threads included choosing Telegram or Discord for an Arabic community, learning programming, AI, and DevOps through practical projects, data residency, hosting and DNS, plus ideas for automating marketing and short-form clip extraction. A recurring conclusion was that building a product does not automatically solve distribution; multi-platform expansion, the pre-launch strategy, Hermes adoption, and parts of the website and domain work remained open or under investigation.
123 moments from the stream
- 6:06:10
From Shopify-First App to Multi-Platform Service
An early strategy discussion considers keeping Shopify as the initial and primary platform while exploring a more independent branch that merchants could use from multiple platforms. Similar services, pricing, and one service’s launch date were examined, but the scope and feasibility of expansion remained under investigation.
- 6:13:38
Switching to Mac: Allowing Time to Learn the System
A discussion about buying a Mac for work and why first-time use can feel difficult. The speaker argues that the system needs an adjustment period and describes his comfort with Mac and iOS as a personal preference shaped by long familiarity, while also liking Windows and rejecting team-like rivalry over platforms.
- 6:17:48
Platform Tribalism and an Opinion on Apple Pricing
The speaker criticizes turning iOS, Android, or phone preferences into tribal loyalty, joking that some people around him treat buying an iPhone like betrayal. He also offers the opinion that Apple sometimes overprices its products.
- 6:23:51
An Impression of GEO Company Growth and Pricing
The speaker says he has noticed GEO companies growing quickly and considers one platform’s pricing expensive.
- 6:25:34
A Brief Opinion on Codex Speed Versus Claude
The speaker describes Codex as able to keep working without tiring, but considers it slower than Claude. This is a brief personal impression, not a documented benchmark.
- 6:27:52
Learning DevOps Through Building and Practice
Responding to a specialization question, the speaker argues that learning DevOps does not necessarily require a long course: examine what practitioners build, create a similar project for hands-on experience, use a few videos when needed, and enlist AI as a learning aid. He qualifies this as his personal view and says he does not know the exact DevOps curriculum in detail.
- 6:30:39
Impressions of GEO Platform Websites
The speaker gives an impressionistic comparison of GEO platform websites, praising Writesonic’s site while finding Profound’s site heavy, and notes that this is not his main field. The captions do not describe a structured performance test.
- 6:31:47
Learning DevOps by Building
Although the speaker does not consider himself a DevOps specialist or come from a programming background, he suggests that building real projects directly is the best way to learn, while a full course is also an option. He also recommends seeking relevant communities and general guidance while avoiding claims of expertise.
- 6:36:42
Comparing Usage Limits in AI Coding Assistants
The speaker describes his impression that Codex sustained a long sequence of coding tasks before reaching a usage limit, unlike another assistant where limits appeared sooner. He explicitly remains unsure whether the difference is temporary or related to free-token policy.
- 6:39:58
Turning Competitor-Platform Inspiration into Product Ideas
The speaker discusses a website he admires and views as closer to a full platform than a conventional site. He thinks a comparable backend might create a competitive opening, while acknowledging that the established product is far ahead and that a calm brainstorming session is needed to choose a realistic development path.
- 6:42:21
Automating Hermes Agent Through a Command Center and Telegram
A discussion of placing Hermes Agent in a command center to handle automation tasks, with remote interaction through Telegram. A GitHub repository was expected to be shared later to explore what the agent could do, but the captions do not establish the implementation details or outcome.
- 6:50:00
Testing a Reference Website as Context for Codex
An experiment explored a practical way to show Codex a reference website so it could describe the page and draw ideas from it. The captions indicate attempts to open and interact with the site inside the working environment, and clarify that the test concerned Codex rather than Cursor, but do not conclusively establish what it could see or extract.
- 6:54:43
GEO and AEO Compared with Traditional SEO
An explanation contrasted traditional SEO with Generative Engine Optimization (GEO), while presenting Answer Engine Optimization (AEO) as a closely related label. Referring to his reading of Google Trends, the speaker said GEO interest began emerging around mid-July 2025 and attributed to market experts the view that discovery through generative assistants may grow, while stressing that the shift remains gradual and its future is uncertain.
- 6:58:27
Observing the Limits of Recreating a Reference Effect with Codex
The speaker was impressed that Codex reproduced an unclear element closely, while noting that a particular effect was not implemented because the agent did not know what that effect was. He said the workflow had become clearer and offered a personal impression that token use was low; the captions do not identify the element or provide a measured usage comparison.
- 7:03:51
Recreating a Web Component’s Visuals with Codex
The speaker describes selecting a component from a website and sending it to an agent for recreation. In his experience, the visual appearance can be captured, while animation is not inferred and must be added manually.
- 7:06:38
Long Streams as a Deep-Work Constraint
The speaker says long streams reduce opportunities for distraction and turn the session into sustained work, but they also make short breaks and rest harder, so he plans to pause before continuing.
- 7:12:37
Choosing Automation for Organic Marketing
The speaker considers trying an automation tool because another option appealed to him but has a learning curve. His practical motivation is an upcoming need to automate marketing work, especially organic marketing; the tool names are unclear in the captions.
- 7:17:06
Milestone Audits to Control Bug Accumulation
The speaker explains that an agent-driven workflow accelerates task completion but can accumulate bugs. He therefore plans to audit each milestone with a dedicated command and checklist, while expecting remaining defects to require a separate quality-assurance phase.
- 7:18:42
A Personal Take on Codex Usage Limits and an Untested Comparison
The speaker says that, in his personal experience, Codex supported parallel work and research sessions without quickly reaching the usage limit. He strongly favors it over Claude, but explicitly says part of that comparison was not tested, so it remains an opinion and prediction rather than a direct benchmark.
- 7:22:56
Exploring Hermes for Easier Automation
Luqman discussed Hermes as a possible option for strong automation with a lower learning curve, asking about its official site, subscription or API access, and local versus VPS deployment. Security, setup effort, and comparisons with another tool remained unresolved, and he said he would research it further before relying on it.
Got a question about something in this stream?
Ask in the group