Talkstore Begins Now! 🟢 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 documents the public start of building Talkstore: a proposed Shopify-specific AI assistant through which merchants could manage products, storefront work, and commerce intelligence conversationally. Its capabilities and possible expansion beyond Shopify are presented as conditional directions, not validated outcomes. Practical work covered Shopify app and development-store setup, Supabase integration and Prisma migrations, secret handling, and repeated diagnosis of session, connection, preview, and Railway deployment problems. The preview and some setup steps worked, and the service later appeared active, but several database, network, build, and deployment issues remained intermittent or unresolved in the covered segments. This led to useful lessons about starting with logs, documenting a stubborn problem, seeking a fresh diagnostic perspective, and reviewing coding-agent compromises rather than accepting a result merely because it works temporarily. The stream also presents a multi-agent command center that separates research and task preparation from implementation, code, security, and compliance review, using dependencies, an adaptive timeline, and human approval gates. Luqman felt this structure accelerated and preserved the context of the work, while still questioning whether every agent was running and visibly logged. He emphasized that authentication, secrets, and nuanced decisions continue to require human intervention, and that task count is a poor measure of difficulty. Work later shifted toward pre-launch distribution: a simple but polished Astro waitlist site kept separate from the Shopify app, with email capture, analytics, social metadata, a teaser, and a distribution campaign. The first landing-page draft had missing hero content and styling or rendering failures, so it was treated as an unfinished iteration. Other useful discussions included audience questions about using AI to explore income paths from existing expertise, SaaS unit economics, choosing between ads and SEO/GEO, Arabic-market WhatsApp automation, and targeting external markets, plus local voice dictation, content automation, and Telegram community options. Stream lag and interruptions were also investigated without a conclusive cause, while music licensing and gym culture appeared as brief secondary discussions.
121 moments from the stream
- 5:56:17
Moving to the Next Milestone While Managing Energy
Luqman says the first Talkstore milestone is complete and considers using his current momentum to begin the next part, while noticing his energy declining and emphasizing eating and drinking properly before continuing.
- 5:57:32
Reviewing pre-launch tasks and removing blockers
Luqman reviewed the first week's milestone progress and asked Claude to investigate why the distribution-campaign and landing-page/waitlist tasks were blocked, unblocking anything ready for execution. He noted that the next work would happen outside Shopify, then reviewed the task board, including securing a domain and deploying the landing page.
- 6:04:01
Briefing a distinctive waitlist landing page
Although functionally simple, Luqman stressed that the page would not be a one-shot task: it should draw on best-practice research, use the brand architecture, logo, and colors, and add motion and effects without feeling generic. The intended result is an appealing but simple waitlist experience with space for a teaser video; the captions end before the final element beneath it is explained.
- 6:07:23
Turning Research into a Builder-Agent Prompt
The workflow delegates exploration and research to agents, then consolidates the findings into a substantial Markdown prompt that guides a builder agent and identifies the skills, plugins, and agents needed to meet the intended standard.
- 6:10:50
Scoping a Simple Waitlist Landing Page
The landing page is intentionally scoped as a simple waitlist, with email collection, possible Supabase storage, analytics options, and Open Graph metadata for presentable social sharing. The distribution campaign is treated as separate from building the application itself.
- 6:15:07
Treating the Builder Agent’s Output as an Iterative First Draft
The builder agent’s result is framed as a first pass rather than a one-shot final implementation, with review and follow-up changes expected. The prepared prompt includes a dark visual direction, restrained motion, and animated interface elements, though some design wording is unclear in the captions.
- 6:18:14
Separating Talkstore’s Landing Page from the Shopify App
A discussion of building the public marketing page with Astro instead of placing it inside the Shopify React Router app. The proposed structure keeps the landing page and app code as separate projects in the same repository with independent builds, rendering most sections as static HTML while hydrating selected interactive sections.
- 6:20:27
Breaking a Large Landing-Page Task into Agent Work
A practical reflection on giving a coding agent a large specification for a complete landing page, expecting the task to take time and require later review. The work was organized across multiple instruction files and divided into several tasks, with Codex mentioned as a possible tool for later UI polishing.
- 6:24:47
An Agent for Reviewing Shopify App Compliance
While reviewing the roadmap and app-submission checklist, a dedicated compliance agent was described that would inspect the codebase against Shopify’s checklist. The review was characterized as complex, and the captions do not establish that the check was completed or passed.
- 6:28:04
Verifying Post-Build Compliance-Agent Execution
Luqman suspected that the compliance agent was not running after builds as intended, because an entire milestone containing tasks and submission checklists had passed without clear evidence of its execution. He requested verification of the execution path, noting that the validator may skip tasks with no checklist items.
- 6:30:55
Embedding Talkstore in Shopify and Comparing It with Sidekick
Luqman examined running Talkstore inside a Shopify app, explaining that configuration had not yet begun and that the product was not intended to be merely a thin model wrapper. He positioned it as a direct competitor to Shopify Sidekick and noted that he had discussed their differences during the stream.
- 6:31:19
Prioritizing the Teaser and Pre-Launch Marketing Campaign
Luqman said the teaser video and pre-launch marketing campaign needed to be completed quickly as he considered the upcoming work.
- 6:32:45
Using a Command Center to Accelerate and Organize Work
Luqman said the command center had practically helped him complete work quickly and in an organized way, while recording the workflow and preserving a clear history. He framed this as an operational benefit beyond a personal preference for organization.
- 6:36:21
Why the Livestream Had a Delay
Luqman explained that the livestream lagged by roughly 30 to 60 seconds because it used the normal-latency setting, accounting for the delay between speaking and viewers receiving it.
- 6:38:33
Tuning a Command Center for Domain-Specific Automation
The speaker explains that the system is more than added code: it organizes automation around the application’s domains. He distinguishes unfinished application-level optimization from tuning the command center to respond to commands and route them by tags, while arguing that practical usefulness is the key criterion.
- 6:42:57
Voice Dictation and Local Speech-to-Text
The speakers discuss voice dictation for turning speech into cleaner written text and giving AI tools longer textual context instead of relying only on a short prompt. They also note that a similar workflow could be built with Whisper running locally, while some technical details remain unresolved in the conversation.
- 6:57:26
Does Ego Dominate Gym Culture?
The speaker begins describing his personal experience with gyms and working out, saying he did not personally feel dominated by ego, before adding a qualification whose ending is unclear in the available captions.
- 7:04:14
Inspecting the Landing Page and Documenting Rendering Issues
He reviewed the landing page and found that sections existed but their styles were not loading correctly, while the hero section was missing. He gathered screenshots to document the problems and noted a missing project configuration that could affect the tool’s behavior.
- 7:06:49
Exploring Automation for Content and Social Media Tasks
He discussed reducing the time spent on short-form videos and social media management by exploring workflow automation instead of relying entirely on manual work. This remained an idea requiring research, not a completed implementation.
- 7:07:46
Revising the Teaser and Distribution Campaign Timing
He revised the schedule because implementation might finish sooner than expected, proposing that the teaser be prepared near the end of the week while the distribution campaign timing is adjusted. He qualified the plan by noting possible future obstacles, including theme-duplication and compliance concerns.
Got a question about something in this stream?
Ask in the group