Inside the Core 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 centers on an in-development Shopify app intended to improve stores’ visibility on AI platforms, with a capability audit exposing gaps between the published plan matrix and actual implementation—especially APIs, webhooks, and multi-store agency support. Luqman demonstrates a review workflow combining specialist agents for security, compliance, and UX with hands-on testing, human-controlled merges, and automated deployment; mobile inspection revealed visual defects, confirming that the product was not finished. The discussion also covers free-versus-paid report design, A/B testing the placement of email capture, and an audit-led acquisition funnel whose channels should be measured before ad spending. Audience questions explore Claude, Codex, agent skills, and conversation-context compaction, with advice to master reliable tools instead of chasing every new model. On GEO, the speaker distinguishes it from SEO while leaving the effects of AI-generated articles, backlinks, and llms.txt unresolved because the evidence is not conclusive. The practical conclusion is to align public claims with real capabilities, understand integrations, and test the complete experience before Shopify submission; the completion date, best conversion flow, and need for external monitoring remain open.
109 moments from the stream
- 0:45:33
Avoid Chasing Every New AI Model
Drawing on his experience during 2025, the speaker argues that switching subscriptions and tools whenever another model briefly pulls ahead creates distraction. His advice is to stay with a reliable model that fits the workflow because model leadership changes continually.
- 0:46:58
How Core Engine Helps Shopify Stores Improve AI Visibility
The speaker gives an early overview of an in-development Shopify app that measures a store’s visibility on AI platforms and applies content improvements such as text, blog material, product descriptions, and FAQs to help that visibility grow over time. He says the app includes multiple agents and recurring updates, but it was still being debugged and was not ready for a full internal demonstration.
- 0:51:23
Avoiding Hasty AI Model Switching
The speaker recalls frequently jumping between AI models in 2025 and says he has become more inclined to wait before changing tools unless a new model shows a large, clear advantage. He adds, however, that an exceptionally high benchmark result for a new model could prompt him to subscribe immediately.
- 0:53:00
Reviewing the User Journey Before Building
Before implementing anything, the speaker asks for a concise restatement of a proposed option and a practical walkthrough of someone arriving from a Facebook ad, covering the full journey and recommendations first. The option or project name is unclear in the captions.
- 0:53:56
Reducing Distraction from AI Tools and News
The speaker argues that the daily flood of tool and model videos can create enough distraction to prevent real work. His practical approach is to focus on one project and apply marketing skills while checking industry news periodically instead of following it every day.
- 0:55:07
Comparing Coding Usage Headroom
Responding to a comparison between OpenAI and Anthropic, the speaker says the issue is not simply generosity and tentatively suggests that OpenAI offers more room for coding work. He defers a firmer judgment until after May 31; the captions do not establish the year or precise usage limits.
- 0:56:20
Choosing Agent Skills and Building Custom Ones
The speaker discusses coding-agent skills, including one aimed at adapting writing patterns for clients and another intended to improve factual precision at a possible additional resource cost. He prefers creating custom skills when the goal is clear and shaping them around relevant frameworks and metrics, while noting that some packaged skills or plugins offer capabilities that are difficult to reproduce.
- 0:59:46
Waiting for Benchmarks Before Judging a New Release
The speaker reacts cautiously to an unspecified version-number jump and postpones judgment until benchmark results are available. The captions do not clearly identify the product or model.
- 1:03:08
Finding Ready-Made Marketing Skills
The discussion clarifies that the mentioned marketing skill was tailored to a particular setup, then demonstrates searching a large skills marketplace for alternatives such as marketing psychology and idea-generation skills.
- 1:05:18
Using AI to Evaluate a Landing Page
A practical suggestion to give an AI model a landing page, request a score out of ten, and ask for areas of improvement, alongside a reference to prior experience using an AI coding tool.
- 1:10:03
Comparing AI Coding Tools
A provisional, personal comparison of Codex, Claude, and another tool whose name is unclear, implying that the choice depends on observed performance and practical resource use and that testing was not yet complete.
- 1:12:01
Differentiating Free and Paid Audit Reports
A review of an audit product’s offer: what appears immediately in the free report and what could justify the full version, such as downloadable detail and granular scoring. It also asks whether paid sections unlock automatically after purchase, whether another scan is required, and whether the proposed value is compelling. The spoken price is inconsistent in the captions, so no exact amount is asserted.
- 1:17:53
Use Caution When Installing Agent Skills
A brief warning not to install every discovered skill indiscriminately, but to select skills carefully before adding them to a working environment.
- 1:24:44
Outbound Outreach and Platform Automation Risks
The speaker says their experience is more focused on digital marketing than traditional sales and suggests that learning about outbound outreach may help them, while warning that platforms such as Reddit may detect AI-automated activity and ban accounts.
- 1:29:02
Pre-Submission Integration Review for Multiple Shopify Stores
A required pre-submission technical review covers scopes, secrets, and webhooks, followed by an explanation of how agencies can use APIs and webhooks to connect a scanning system across multiple Shopify stores. The API and webhook work is described as one of the harder parts of the current milestone.
- 1:33:41
Defining the Audience for a Shopify Tool
The speaker explains that the tool under discussion is primarily intended for merchants with a Shopify store, so it may not be useful to people who do not operate one.
- 1:34:50
From SEO to GEO and Comparing Tool Limits
The speaker praises Neil Patel’s SEO expertise and considers the growing interest in GEO useful. He separately says the tool is not really a competitor and mentions search and monthly prompt limits. This is the speaker’s assessment, not a documented test result.
- 1:36:29
Starting an E-commerce Business with Shopify
The speaker presents Shopify as a way for someone with products to build a website and begin selling quickly, while noting the importance of website traffic.
- 1:38:11
Do AI-Generated Articles Help GEO?
The speaker discusses a feature that generates AI articles within Shopify, but says his research has not yet found conclusive evidence that it improves GEO, leaving the question open for further investigation.
- 1:39:25
Prioritizing Product Capabilities with a Matrix
The speaker describes using a capabilities matrix to compare the product with larger tools, selecting feasible features and excluding impractical ones while balancing value against operating cost.
Got a question about something in this stream?
Ask in the group