Building the Launch Machine 🔴 Building to $100K
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 building TalkStore, envisioned as an AI agent that takes practical actions in Shopify stores, while preparing for launch early. Luqman discusses using a landing page, waitlist, distribution campaign, and organic marketing to test merchant interest before the product is finished, while noting that sign-ups do not guarantee conversion. He iterates extensively on the landing page: documenting its visual direction, handing specifications to agents section by section, inspecting reference animations and assets, and diagnosing gaps between agent-reported changes and browser results. Some motion eventually appeared, but the design, distribution work, and teaser video still required follow-up. He also develops a multi-agent workflow for research, building, review, security, and compliance, choosing task-scoped Shopify documentation retrieval through Context7 over full ingestion without independently validating its accuracy. Other useful discussions cover automating Shopify submission audits, designing a multi-intent message classifier, resuming work after session limits, and audience questions about frontend, backend, and hosting tools. Open questions include the reliability of automatic agent-chain dispatch, completion of the reviews, the landing page’s final direction, and the distribution-automation plan.
66 moments from the stream
- 2:25:42
Reviewing the Launch Command Center and Building Page Sections
A practical launch-interface review identified overlap between distribution nodes and the checklist, plus a missing waitlist-tracker bar, while also checking that the app runs from the correct folder. Work then moved into building page sections, treating the third section as central and prioritizing visual effects before colors, typography, and content.
- 2:32:48
Diagnosing a Viewer Counter in a YouTube–OBS Integration
A viewer counter that was not updating correctly prompted a short diagnostic sequence: open the correct app, refresh or reload the browser element, and consider resetting the YouTube–OBS connection. The diagnosis was not pursued to completion.
- 2:33:42
Improving Launch-Diagram Labels and Merchant-Bar Scheduling
The review proposed moving three node labels to the side to avoid colliding with the merchant-bar label, and tying the bar to the launch schedule so it extends at least two weeks ahead of launch.
- 2:35:33
Framing the Problem Section for TalkStore
The page drafts a problem section around obstacles Shopify merchants face when searching for tools or hiring freelancers and waiting through revision cycles, positioning TalkStore as a response. Some displayed wording remained rough or unclear.
- 2:39:19
How to Describe a Livelier Interface to a Coding Agent
He critiques the page as flat and static, asking for an atmospheric background, cursor interaction, product-related motion, and a lighter alternative to the familiar dark aesthetic. His strategy is to begin with deliberately strong animation and tune it down later, while first asking the agent for ideas rather than immediate execution.
- 2:40:58
Practical Streams and an Arabic Vibe-Coding Community
He says live streams can demonstrate practical building in public more fully than simple videos, and suggests that audience interest could eventually support an Egyptian or broader Arabic vibe-coding community.
- 2:46:51
Turning Visual Direction into a Prompt and Reviewing the Result
He selects a highly animated direction, transfers it through a prompt file for the agent to apply to the hero, and reviews the preview. The result still feels more static than intended; reference sites and effects such as subtle smoke inform the exploration, but he continues seeking stronger motion that can later be reduced.
- 2:53:15
Collecting a Reference Page’s Structure and Assets for Hero Analysis
Explores a practical way to understand a reference page’s hero effects: inspecting the DOM and SVG files, requesting beginner-friendly steps, considering Firecrawl, then saving the complete webpage to collect HTML and related assets; SingleFile was also mentioned as an alternative extension. The captions do not establish that Firecrawl was used successfully.
- 3:00:38
Breaking Down Cinematic Web Effects into Video and SVG Assets
Explains why matching a high visual standard is harder than assembling a basic landing page, then reports an analysis concluding that the reference smoke effect is a pre-rendered video while the light beams are SVG images rather than CSS gradients. These are findings reported in the discussion, not independently verified tests.
- 3:01:38
Fixing UI Layer Order and Trying a Cursor-Following Glow
The UI review asks for a bar to sit above vertical guide lines, explores a call-to-action button with a warm glow that follows the mouse, and notes that the update still did not appear after reopening the app. The captions do not confirm a successful fix.
- 3:09:09
Launching Before Full Polish and Building a Waitlist
Discusses postponing some visual polish once the page is good enough to launch, since further refinement may be less important before there is traction. Paid advertising and organic growth are raised as possible ways to build a waitlist, with organic growth described as important.
- 3:10:45
Diagnosing a Persistent UI Issue After Relaunch
Uses a screenshot to show that a vertical line still appears above the waitlist bar after quitting and relaunching the app, and asks why the problem persists. Later, the atmospheric background effects become more visible while the hero headline becomes invisible, revealing a new visual regression whose cause is not established.
- 3:14:34
Positioning TalkStore as an AI Agent for Shopify Stores
TalkStore is described as an AI agent that reads a Shopify store and takes actions such as generating sections, enriching products, and fixing problems through conversational interaction.
- 3:15:08
Why Landing-Page Design Needs Focused Off-Stream Work
The discussion explains that refining the landing page requires visualization, sustained internal focus, and quiet time before forming the right prompt, so the work was deferred off-stream rather than presenting a long, low-interaction process to viewers.
- 3:18:52
Reviewing the Command Center and Planning the Move to Week Two
The command center showed the project still in week one, prompting consideration of moving directly into week-two tasks while the website, distribution campaign, and teaser video remained items requiring follow-up.
- 3:22:21
Preparing Development Tasks with Specialized Agents
Week-two execution began with application development prioritized and landing-page work deferred off-stream. A task is sent to research and explorer agents; the researcher checks documentation when needed, their findings feed a prompt file for the builder agent, and the agent hub displays their activity and timing.
- 3:27:13
Design Decisions for a Message Intent Classifier
The task preparation covered a classifier that maps messages to a domain, verb, and confidence score. The design favors an ordered array with a primary intent, a database-backed prompt configuration, persistence of the reasoning field alongside messages, and a rough input-token estimate for possible later billing logic. It also notes that Vercel AI SDK v6 replaces a deprecated approach with text generation using object output.
- 3:29:04
Development Tools for Backend and Frontend Work
Answering an audience question, he said he uses Warp for most backend and application-building work, including multiple terminals, and switches to Cursor for UI work.
- 3:33:01
Alternative Website Hosting Options
He suggested Cloudflare and Railway as hosting options, describing Cloudflare as somewhat easier, suitable for websites, and relatively similar to Vercel, while calling Vercel one of the leading choices. These were his personal recommendations in an audience discussion.
- 3:35:15
Prioritizing the Product Before the Landing Page
He felt landing-page design requires visual thinking and back-and-forth iteration that is difficult for him on-stream. His preference was to finish it separately, focus first on building a strong product and showing its progress and command center, then move toward launch and marketing.
Got a question about something in this stream?
Ask in the group