New SHA8AL.com + Sarmad
Stream summary 75This stream in minutesWant 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 centered on turning SHA8AL.com into a community hub and searchable bilingual knowledge base that links topics and tools to timestamps across the broadcast archive while showcasing members’ projects. Luqman discussed an automated workflow for downloading YouTube captions, filling gaps locally, and then using models to summarize, select passages, and produce short and longer videos. Arabic transcription accuracy, selection criteria, video quality, and archive-processing cost still required evaluation, and no completed system was demonstrated. Practical work included reviewing Burnish and smoke-testing its installation and free-plan flow, with proposals to reduce onboarding friction, clarify what counts as an install, improve the analytics dashboard, and track Shopify App Store visits. A recurring conclusion was that distribution and traffic—not product construction or the funnel alone—were the main bottleneck, prompting exploration of measurable channels such as outbound outreach, agency partnerships, advertising, and communities, without claiming final results. The stream also covered Sarmad as an automation graph system, Paperclip and goal-directed agent coordination, build-versus-buy tradeoffs, and privacy and usage-limit constraints. Luqman offered qualified, non-benchmark impressions of Codex, Claude, GPT, transcription, and video tools, emphasizing hands-on testing, adequate context, and validation when model outputs conflict. Audience discussions additionally touched on building a Micro-SaaS with AI, market choice, freelance pricing, AI’s impact on jobs, and automated social replies, while several proposed workflows, comparisons, and tests remained unresolved.
177 moments from the stream
- 2:12:41
Automation Tools for Repetitive Tasks and Data Extraction
The speaker discusses using tools for data extraction, automation, and repetitive loops, and says he preferred GPT automation in his experience because another tool encountered a Chrome connectivity problem.
- 2:16:30
Comparing a Caption Downloader with the Library Workflow
A comparison between a notebook that downloads YouTube captions and a broader workflow that adds coverage checks, verification, and editorial work. The shortcut may save time, but the discussion concluded that the main approach already includes similar functionality and does not need to be replaced.
- 2:18:43
Separating Local Automation from Analysis Tasks
The workflow separates scriptable tasks, such as transcript handling and reframing horizontal video vertically, from judgment-heavy work such as reading captions and selecting suitable passages. Selecting complete, appropriate passages from Arabic captions was described as the current bottleneck.
- 2:20:29
Defining Clear Criteria for Passage Selection
The discussion asked what criteria the workflow should use to select passages and how its system instructions should guide that decision. The exact local transcription implementation remained uncertain at this point.
- 2:22:29
Downloading YouTube Captions and Filling Gaps Locally
The workflow was clarified: scripts download YouTube captions directly, while missing or incomplete material can be transcribed locally with Whisper using the computer's processing power. Model usage comes later for reading, summarization, and passage selection that require analytical judgment.
- 2:25:54
Indexing More Than the Work Performed
The library should not index only the work performed. General topics, audience conversations, and question-and-answer discussions can also be useful, while practical work remains important.
- 2:26:35
YouTube Growth Difficulty and Audience Affinity
The speaker's view was that YouTube is harder to grow on than TikTok and Instagram, but may produce stronger audience interest and affinity. This was presented as an opinion, not a tested result.
- 2:27:35
Smoke-Testing the Burnish Installation Flow
A practical smoke test began for installing Burnish and choosing its free plan, walking through the initial screens and expected states. No clear issue was observed in the first two steps before attention moved to the plan-selection screen.
- 2:29:41
Reducing Friction in Burnish Onboarding
The Burnish onboarding was proposed to be reduced to one page or to take merchants directly into the app, because every extra step adds friction and can cause users to drop off. Consolidating the essentials into one step was suggested to reduce abandonment before reaching the application.
- 2:31:27
Simplifying Onboarding Through the Free Plan
The review found that continuing on the free plan works even though its card cannot be clicked, while payment through the paid-plan cards had not yet been verified. The speaker considers reducing onboarding to one step or sending users directly into the free plan, then presenting strong upgrade incentives inside the app.
- 2:33:34
Using a Different Model for Strategic Questions
The speaker says they prefer turning to a different model when seeking strategic judgment, viewing the quality of an answer to “what do you think?” as potentially more important than coding ability in that context. The model identities are unclear in the captions.
- 2:34:16
Connecting Product Improvements to Shopify Search Positioning
The discussion turns to completing a section referred to as “Layla,” then showing the product’s Shopify search position for selected industry keywords to assess potential improvement. The implementation strategy is not yet established in this passage.
- 2:37:38
Direct App Entry Instead of Multi-Step Onboarding
The speaker contrasts the proposed direct-entry flow with an earlier 3–4-step onboarding flow adopted after reviewing competitors, and notes that the change would require additional engineering work.
- 2:39:45
What Should Count as an Install?
An unresolved discussion about defining the install event. One proposal keeps the regular install definition because it is simpler even with free entry; a stricter alternative would count a store only after the engine’s answers arrive and the merchant returns to open the app.
- 2:40:35
Distributing Upgrade Offers Across Plan Tiers
A proposal for the in-app upgrade offer to cover every plan tier rather than a single plan, using hypothetical incentives that vary by size or duration and make higher tiers more attractive. The details were exploratory examples, not a finalized decision.
- 2:53:02
Install Attribution Timing and Q4 Offer Messaging
A discussion of delayed install attribution when a merchant opens the app on a later day, alongside the point that a strong offer still needs clear communication and sufficient traffic. It also suggests cautiously using the Q4 2026 and Black Friday context without overdoing the messaging.
- 2:55:15
Reducing Onboarding Friction in Free Apps
Onboarding is presented as a product choice, not a requirement for Shopify apps. The stated view is that a free app may benefit from sending users directly into the product and offering upgrades inside it, while traffic acquisition remains more important than polishing surface details.
- 2:57:40
Growth First: Outbound Outreach and Agency Partnerships
The discussion prioritizes users and installs before optimizing monetization, while inviting business-partner-style pushback. The distribution approach combines outbound outreach to Shopify agencies that can introduce the app to client stores with inbound posting, while acknowledging that traffic remains limited.
- 3:00:38
Improving Sarmad Analytics and Tracking the Shopify Listing
Requested Sarmad improvements include showing traffic counts when hovering over daily-trend bars and making the results refresh state cleaner. The discussion also explores tracking visits to the Shopify app-listing page through GA4 or possibly server-side Google Tag Manager, but leaves the method unconfirmed and requiring investigation.
- 3:05:29
The Meaning of Sarmad and the Challenge of Naming a Business
“Sarmad” is explained as a term conveying permanence or eternity and as a word also used in Sufi contexts. The discussion then notes that naming a public business is difficult because domain availability places an important constraint on the choice.
Got a question about something in this stream?
Ask in the group