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
- 2:53:34
An AI Agent That Analyzes Shopify Product Data Before Acting
A prototype agent is tested on a Shopify store to inspect products, assess data quality, identify missing fields, and propose product enrichment. The discussion emphasizes that its key value is taking store actions rather than only understanding the catalog, with a confirmation gate requiring user approval. The gate is shown working, while the basis for some apparent image assessment remains unclear.
- 3:03:47
Debugging Agent Action Execution
A debugging session investigates a failing action in the agent prototype. A targeted test is used to distinguish a server-runtime problem from an issue in the store backend; after the runtime path appears unlikely, attention shifts to a pipeline with heavy stages that may fail silently. The priority is validating and fixing the action logic before polishing the interface.
- 3:14:30
Finish the App Quickly, Then Polish the UX
The speaker prefers reaching a ready version of the app quickly, then spending separate time polishing its interface and user experience before release; the stated timing is an estimate, not a confirmed result.
- 3:15:59
Adding Error Logging to Locate a Problem
The speaker explains that the implemented change added error logging to identify the source of a problem, while also noting a UX issue that still needs attention.
- 3:19:46
A Browser-Assisted Workflow for a Coding Agent
Anticipating repeated back-and-forth with his coding agent, the speaker proposes a dedicated workflow that lets Claude access the browser, inspect the conversation or interface, and capture a screenshot instead of requiring manual image sharing each time. There is also a brief reference to a conversation handler and basic prompts.
- 3:21:52
Audio Comfort and Product Details
Responding to an audience question, the speaker agrees that phone audio should be comfortable to listen to and says details of this kind should be available on the Amazon product page. Parts of the original question are unclear in the captions.
- 3:24:09
Debugging Product Data Enrichment with Classifier Logs
An interactive debugging pass on a function that fills missing product fields, using classifier logs to inspect the detected domain and resulting decision. The request is then corrected to focus on editing the snowboard length rather than a metafield. The discussion notes that even one function can require extended debugging and concludes by reviewing a classifier confidence score of 0.92, while some implementation details remain unclear in the captions.
- 3:35:49
Using Milestones to Reduce End-of-Session Workload
The speaker prefers completing and handing off work at each milestone instead of accumulating everything until the end, reducing the final workload.
- 3:39:39
Investigating Logic Errors Instead of Guessing
While tracing a problem with no clear error appearing in the log, the speaker wants the coding agent to stop attempting repeated small fixes and investigate the codebase to understand the source of the logic errors instead of guessing, noting generally that this investigation consumes usage tokens.
- 3:44:25
Debugging an Automated Product-Description Rewrite Action
The speaker traced a failing action intended to enrich or rewrite a product description, returning to the error log and requesting a line-by-line review of the context and codebase without breaking the code. The captions appear to indicate an issue involving an input argument, followed later by an observed description rewrite and confirmation that the action could execute. Shopify appears in the inspection context, while the exact field details remain unclear.
- 3:59:20
Planning a Focused Debugging Session Without Rushing the Project
The speaker said the remaining debugging would require time and preparation. He wanted to finish and submit the project without rushing or compromising the work, then revisit other ideas mentioned earlier in the stream.
Got a question about something in this stream?
Ask in the group