BUILDING ≠ SELLING
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 balancing Shopify app development with early marketing. The speaker identifies the landing page and distribution as the current bottleneck, then works through a simpler site narrative, trust and waitlist conversion, section reviews, branding, logo variants, and social previews. A test sign-up was confirmed in Supabase and the OG image was reported as serving correctly, while Umami conversion-goal setup and analytics testing remained unfinished because of unreliable interface behavior. Because the guidance appeared to reference an older interface, research was requested for the UI available on April 2, 2026 before proceeding. The stream also presents a multi-agent workflow for task management, research, and building, alongside a theme-aware Liquid section-generation pipeline that validates syntax, JSON structure, and CSS scoping. A fragmented discussion proposes feeding validation errors back to the AI for another attempt, but does not establish a completed implementation or test. Shopify compliance rules, its distinct billing path, and post-deployment reviews are presented as launch requirements that belong in the implementation plan. An independent reviewer is discussed only as a possible later addition that could catch issues missed by the builder; the design of specialized review and security agents still requires research, and shared infrastructure for an app suite remains unresolved. In audience discussions, the speaker offers his personal experience of starting with a small toolset and a useful, enjoyable project despite lacking a strong programming background; he also explains short-form clip automation, context handoffs, and practical usage-limit management. Comparisons among tools and models are framed as preferences or uncertain observations, with benchmark rankings treated cautiously. Generated imagery, animation, multi-shot video, and an FFmpeg alternative are explored without claiming completed experiments, while extending Talk beyond Shopify is raised only as a future feasibility question.
59 moments from the stream
- 0:53:49
Umami or GA4 for a Simple Landing Page?
The speaker prefers Umami for this case because the needed signals are limited to traffic sources, device type, on-site behavior, and waitlist conversion. In his view, GA4 is better suited to more complex analytics such as ecommerce funnels, conversion rates, and add-to-cart activity.
- 0:56:13
Start with a Fun Personal Project
The speaker recommends starting with a fun project that is useful to its maker. After building something useful for oneself, a person may begin considering other things they can create and eventually tools that help others.
- 0:57:15
Basic Umami Setup Steps
The visible setup flow includes adding a website, placing the tracking code in its HTML, and choosing a measurement goal such as views, visitors, events, sign-ups, purchases, or campaigns. Some interface-related wording in the captions is unclear.
- 0:59:56
Narrowing the Agent Team for a Specialized Task
For a task the speaker describes as specialized, he suggests skipping research or explorer agents, using only the builder, and proceeding step by step. The task itself is not clearly identified in the captions.
- 1:01:40
Setting Up Conversion Events and Traffic Sources in Umami
A first-time exploration of Umami focused on creating a waitlist-signup event and conversion goal, measuring conversion rate, filtering traffic by source, and using UTM data. Because the guidance appeared to reference an outdated interface, up-to-date research was requested for the UI available on April 2, 2026 before proceeding.
- 1:05:19
Reviewing Brand Prominence and Dark-Background Logos
A Talk Store interface review flagged weak text alignment and prominence, a less effective compact font treatment, and an unsuitable logo. The feedback called for consulting the existing brand-architecture assets and using the logo variant intended for dark backgrounds.
- 1:05:45
Balancing Interface Polish with Building the Application
A practical reflection that visual details such as colors and presentation matter but had become tiring, prompting a preference to finish the current section and return to the application’s technical and backend work, which still needed more time.
- 1:11:48
Reviewing Logo and Typography on Dark Backgrounds
A practical review of the name placement, typeface, and tagline alignment, noting that an icon intended for a light background is unsuitable on a dark one. The discussion stresses opening the site, taking screenshots, and visually checking the result before treating the task as complete.
- 1:13:14
Setting Up Sign-Up Tracking and Reviewing the Analytics UI
The analytics dashboard is checked for incoming tracking data while a goal event for sign-up or the waitlist is configured. During this session, the interface appeared unreliable and sometimes required a refresh, so further testing was deferred.
- 1:22:08
Verifying Waitlist Sign-Ups in Supabase
He tested the waitlist form, then checked Supabase and refreshed the view to confirm that a record had been created.
- 1:23:16
Build Quickly, Then Reserve Time for Refinement
He described a strategy of completing the core structure quickly while expecting detail work to take additional time. Remaining priorities included refining the landing page and possibly preparing a teaser video before the distribution campaign.
- 1:26:24
Using an AI Usage Window Efficiently
He suggested checking remaining usage before an expected break and running useful work so capacity can refresh while away. It was presented as general time-and-usage management advice, not a guarantee of completed work.
- 1:27:33
Genspark or Claude for Terminal-Based Work?
He viewed Genspark as useful for producing presentations, spreadsheets, and PDFs within its platform, while preferring Claude for building directly in the terminal because it can fit into that workflow. He also said he did not know the other mentioned service well enough to assess it.
- 1:30:02
Adjusting Letter Spacing and Logo Variants
He reviewed the logo, requested wider letter spacing and a background-appropriate variant instead of the white-background version, then prepared light and dark versions for SVG export.
- 1:34:49
Sequencing Dependencies and Running AI Tasks in Parallel
He moved to the task board, caught an error in the starting order, and used an earlier week-and-dependency classification to identify prerequisite tasks. He then ran parallel sessions for preparation work and planned a later generation pipeline while monitoring usage during execution.
- 1:42:18
A Multi-Agent Workflow from Research to Building
A walkthrough of a workflow where a builder agent reviews exploration, research, and task instructions, then handles multiple tasks through preconfigured agents while their progress and logs are monitored in an agent hub.
- 1:43:05
Usage Limits in an Agent-Heavy Workflow
The speaker generally compares usage consumption across AI coding tools and observes that an agent-heavy workflow can exhaust allowances quickly. He suggests that stronger tiers may handle this workload better, while their higher cost creates a practical tradeoff.
- 1:49:29
Designing Problem-Section Visuals with Generated Images and Animation
He describes a setup using Nano Banana for image generation and an animation tool to bring those images to life, with image-based examples for a problem section. He then asks for visual ideas around product operations and commerce intelligence; the resulting ideas are not included on this page.
- 1:50:35
Working On-Stream and Off-Stream
Answering an audience question, the speaker says he tries to work every day and continues off-stream when he is not live, with travel usually being the time when he stops working.
- 1:53:49
Validating Generated Liquid Sections Before Installation
An explanation of a task that acts as a gatekeeper in a Liquid-section generation pipeline, checking the settings or schema block for mistakes before code is installed in a store theme. The review considers error-message clarity, why only three initial checks were chosen, and whether more checks are needed; the stated view was to ship the current version and address additions later.
Got a question about something in this stream?
Ask in the group