PROJECT & VISION UPDATE
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 Talk Store’s change of direction after the speaker says Shopify did not approve the permission needed for its original conversational storefront-editing concept. He therefore reprioritized a faster submission-ready version and deferred broad marketing. For future products, he proposes testing several ideas through landing pages and real demand signals—such as sign-ups, reservations, and conversion metrics—before investing heavily in development; this is a proposed method, not a completed validation result. The practical work demonstrates a multi-agent workflow that turns research into tasks and coordinates building, review, security, and validation, while tracking progress and distinguishing genuinely parallel work from blocked dependencies. Although the speaker believes this can accelerate development, he stresses rapid usage-limit consumption, the need to challenge model estimates, and the importance of human review and UX polishing after agents report completion. The stream also examines a prototype agent that analyzes Shopify product data, proposes enrichment, and requests user approval before making changes. The discussion covers rule-based and pattern-matching quality scores rather than AI scoring, with attribute weights varying by product type; it also recommends choosing suitable model quality before optimizing local-inference costs at scale. The confirmation gate is shown working, but deployment problems, failing actions, and a pipeline that might fail silently require log inspection and code-level investigation rather than guesswork; security review, practical testing, and final polish were not consistently complete. Shorter discussions cover ROI-based hardware choices, coding-tool limits, Soundcore headphones, live-viewer behavior, and the pull of short-form video. In a side discussion, the speaker described Egypt’s marriage “qa’ima” custom as unusual and expressed the opinion that rising divorce has other causes not necessarily tied to it, while deferring a fuller discussion to another stream.
71 moments from the stream
- 1:59:07
Scaling Hardware Around ROI
The speaker advised starting with a Mac mini and moving to more expensive hardware only when there is a clear return on investment, rather than treating a large purchase as the automatic first step.
- 2:04:03
The Pull of Short-Form Video
As a side observation, the speaker said many people have become hooked on short videos such as Shorts and Reels because of their seconds-long viewing pattern.
- 2:04:48
Accelerating App Development with Agents
The speaker expressed the view that ample agent capacity and intensive work could compress the development of a substantial application into a very short period, but presented this as a personal estimate rather than a demonstrated completion rate.
- 2:08:16
Measuring the Value of Extra Usage
The speaker treated parallel work as a practical exercise for measuring how much output extra usage produces, tracking tasks in a command center to compare consumption with results approximately.
- 2:09:50
From Task Completion to Application Polish
The speaker reviewed progress through a block of tasks and expected to finish the remainder soon, while noting that polishing the application might take longer than implementing its core functionality. Some task and feature labels in the captions were unclear.
- 2:11:28
Timing Marketing Around App-Store Review
Discusses a cautious approach of submitting an app to Shopify first, allowing for possible rejection, and postponing major marketing effort and spending until the submission is accepted into the ecosystem. It could then remain a side project supported by organic marketing and limited ads.
- 2:13:11
Prioritize Quality Before Optimizing Local Model Costs
Explains that the value of running multiple models locally depends on how an application uses them and its scale; local inference may save money at high usage, but cost optimization should not come before suitable output quality. The recommendation is to start with quality and optimize costs when scaling.
- 2:17:37
Layered Agent-Assisted Security Auditing
Describes a workflow where a security agent audits code after each step or task, followed by a validator agent as an additional layer after security updates or code review, while final approval remains with the human supervisor.
- 2:22:50
Speed Versus Token Usage in Agent Workflows
The speaker reported unusually fast progress with an agent-assisted workflow, but described high token consumption as the drawback and said workflow decisions must account for current session limits.
- 2:24:20
Planning Agent Work Around Usage Limits and Quality Review
The speaker reviewed the task board and agent workload, anticipating that a later phase would be harder and might need a smarter review-and-refinement workflow instead of repeated back-and-forth changes, while accounting for available usage limits.
- 2:26:24
Shopify App Deployment and Human Verification
The speaker was waiting for a Shopify app deployment task to inspect what had been built, explaining that marked tasks on the board require him to examine the agent's work and understand the blocks it created. Testing and security scanning were pending work, not completed results at that point.
- 2:27:15
Specialized Agents for Building, Review, Security, and Validation
The speaker said Command Center had changed his workflow and that agent configurations could be adjusted for each project. In the example shown, review, security, and validation agents followed the build, and he reported that they found and fixed issues within a workflow tailored to the project.
- 2:33:15
Manually verifying a final milestone task
The task was identified as a human task requiring direct review and verification of the milestone. The working assumption was that it concerned the blocks and did not need the full workflow for preparing an instruction file, launching a builder, and running subagents.
- 2:35:05
Identifying the headphone model
After correcting initial guesses, the headphones were identified as the Soundcore Q30.
- 2:36:40
Inspecting Shopify app block errors before deployment
The inspection surfaced three issues across five blocks. Fixes were started, with the note that the app first needed to be deployed so the blocks could be viewed and tested in practice.
- 2:43:31
Diagnosing Deployment Errors Through Error Patterns
The discussion notes that the first fix failed and recommends understanding the problem and its error pattern instead of guessing, potentially distributing the investigation across agents and checking the correct TOML configuration.
- 2:45:33
A Practical Verification Plan for a Product Ops Milestone
The speaker requests an end-to-end review of the Product Ops milestone after its final task, using practical examples and specific chatbot prompts, while deployment issues involving Shopify metaobjects are being addressed separately.
- 2:47:30
Comparing Soundcore Q30 and Soundcore Space
A brief opinion-based comparison suggests Soundcore Space may be better in some respects, while the speaker prefers the Q30 design and considers its performance sufficient for his needs.
- 2:51:02
Evaluating a Usage-Tier Upgrade From Actual Limits
The speaker proposes a general way to assess an upgrade: observe how often usage limits are reached and whether the current tier obstructs work, rather than deciding from theoretical capacity alone.
- 2:51:36
Testing Product Quality Scoring in Shopify
The work moves to testing Product Ops in the application after the Product Intelligence portion, using a prompt to score product quality and then reviewing Shopify product data such as specifications, organization, price, and metafields.
Got a question about something in this stream?
Ask in the group