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
- 0:02:42
Designing APIs and Webhooks for an Enterprise Tier
The speaker discusses an enterprise tier aimed at companies and agencies, including APIs and webhooks for applying the service across multiple stores, and explains that these requirements introduced more complex technical work than the basic tiers.
- 0:06:28
Auditing and Debugging a Shopify App with Claude Code
The speaker outlines a workflow for reviewing an app that has entered testing: checking onboarding, design, Shopify policy compliance, architecture, APIs, and webhooks while Claude Code performs a supervised walkthrough and debugging pass. He stresses that knowing the purpose of AI-generated components is not enough; their operation should be understood before the app is submitted.
- 0:10:25
Auditing the Capability Gap Between the Website and Code
A review comparing the published capability matrix with the codebase found that a large share of the listed capabilities had not yet been implemented, including APIs and webhooks. This prompted a focus on aligning public claims with what the application can actually do, followed by debugging and filling capability gaps.
- 0:12:25
Short-Form Video as a Discovery and Community Channel
A practical observation that publishing short-form clips across social platforms can help keep a community active and attract attention, while momentum drops when posting stops. Shorts are presented as a discovery and demand-generation channel, not a guaranteed outcome.
- 0:14:01
A GitHub Workflow for Reducing Task Conflicts
The workflow isolates each task in its own worktree, then creates a pull request that is manually merged into the main branch before moving on. Experiments with other tools and models were deferred to reduce distraction until the current product priority is completed.
- 0:15:37
Exploring a Tool for Automation and Customer Service
An initial discussion about learning a tool with a Kanban board and using it to automate application operations, especially customer-service tasks, while retaining direct control of the code. The tool’s identity is uncertain from the captions.
- 0:16:33
Designing an Audit-Led Customer Acquisition Funnel
A proposed acquisition funnel starts with ads and creative content leading to a landing page, where a store URL is submitted to generate an audit report, capture an email, and optimize calls to action. The initial report could lead to a deeper report and a product incentive, but the speaker qualifies this as only one possible strategy and says broader marketing execution is deferred until the application is ready.
- 0:19:49
Automatic Deployment After Pull-Request Merge
A brief deployment walkthrough shows a pull-request merge triggering deployments to Cloudflare and Railway. The page change may take a few minutes to appear after the merge.
- 0:20:39
Keeping Agent Merges Behind Human Approval
The speaker explains that the agent can perform merges, but he requires explicit approval after earlier problems, at least until the workflow is clearer and more systematic.
- 0:22:49
Application Checks and UI Deployment with Claude Code
The speaker describes a workflow he says helps check the database, webhooks, UI, and network, then observes that a deployment completed and the landing page changed after refresh, while noting that it still needs refinement.
- 0:24:50
Starting a Store Audit from a Shopify URL
The segment explains that the substantive audit flow begins after a merchant enters a Shopify store URL and starts the audit, but the following steps are incomplete on this page.
- 0:25:56
Competitive Pressure and Shopify App Growth
The speaker describes a sense of competitive urgency and says he believed others were then building similar Shopify apps. As an indicator of the field’s rapid growth, he claims that 750 apps had been accepted in a single day roughly two weeks earlier.
- 0:27:37
Shopify’s Built-in Suggestions vs. Specialized SEO and GEO Tools
The speaker argues that Shopify offers useful AI-assisted suggestions for product category, type, and alt text, but that these do not remove the need for deeper SEO and GEO apps. Attribution measurement and UTM setup are given as examples of more specialized functions requiring store-side configuration.
- 0:31:55
Designing Free and Paid Reports
A discussion about defining the final product first and then building its stages incrementally, including what belongs in a free report and what justifies a paid report. The proposed price is framed as a signal of prospective-user commitment, with the speaker leaning toward using the existing detailed report as the paid product rather than giving it away in full.
- 0:37:01
Auditing Milestones with Specialist Agents
An explanation of a workflow in which four specialist agents review an entire milestone from perspectives including technical correctness, security, user experience, and compliance. The flow begins with an automated audit, then moves to a shared walkthrough when the review passes, with issues recorded or corrected along the way according to their nature.
- 0:37:56
Opportunity, Work, and Luck
An opinion-led discussion arguing that some opportunities are created through work and initiative rather than waiting for ideal circumstances, while also acknowledging that luck gives some people an advantage.
- 0:39:12
Generating Report Sections and Gating Access
A brief note that six sections are generated for every scan, alongside section-level access gating for paid content. The caption evidence does not fully explain the gating mechanism.
- 0:40:55
Is Break-Even the Measure of a Project’s Success?
The speaker explains that success depends on the project and that revenue need not cover expenses from day one; some businesses may accept a temporary burn rate as an investment intended to produce a later return.
- 0:42:09
Debugging Core Engine with a Coding Agent
The speaker describes focusing on Core Engine and using a coding agent to inspect the application step by step through webhooks, APIs, network activity, logs, and documentation, while reviewing scopes and a capability matrix to identify and follow up on incomplete work.
- 0:43:36
Claude or Codex for React Native Development?
Answering a React Native question, the speaker gives his personal preference for Claude in conversational coding and suggests Codex may suit heavier workloads. When practical, he recommends using both in sequence—for example, planning with one and passing the resulting work plan to the other—while considering cost.
Got a question about something in this stream?
Ask in the group