Building AI Apps Live from Cairo 🔴
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 focuses on methodical preparation for building an application and an AI-agent command center: tracing tasks back to the manifesto and roadmap, auditing them with several language models, and consolidating the reviews into a canonical task file. Questions about timeline inclusion, dependencies, and how best to judge conflicting model reviews remain partly unresolved. Luqman also works through a task-board design that preserves context, run history, acceptance criteria, guardrails, and verification protocols while assembling task-specific agent instructions. Early prompt previews appeared too long and generic and still required refinement and testing; he likewise expected a large automated implementation to need debugging rather than assuming it would succeed. Wider discussion covers live UI previews inside the development environment, parallel work with agent teams, controlling agent scope, validating task integrity before interface polish, and building without a strong programming background or paid third-party tools. The stream also touches on Shopify submission preparation and the need to connect product building with distribution, marketing, and waitlist strategy. Luqman presents these as observations from a public learning process, arguing that logical thinking and a clear workflow can help while leaving both the build and its eventual market results uncertain.
46 moments from the stream
- 1:56:45
Do You Need a Paid Third-Party Building Tool?
He argues that third-party tools may simplify building but are not always necessary when a developer already has a suitable IDE-based workflow; some cited tool names are unclear in the captions.
- 1:58:48
Usage Status Missing in Plan Mode
He observes that plan mode is working while the usage-status indicator is not visible, tests another session, and tentatively attributes it to a system outage.
- 2:00:41
Assembling Prompts from a Single Source of Truth
He reviews a plan to add IPC handlers and an app bridge so the assembled prompt reaches the clipboard and the dispatch is logged, then connects it to task and documentation files under the project root as the source of truth, with cross-references where needed.
- 2:05:09
Expecting Debugging in a Large Task
He expresses clear skepticism that the automated implementation can complete the large task in one pass and expects debugging to be necessary, without claiming success in advance.
- 2:08:42
Checking the Generate Preview Interface
He opens a task and follows the interface instructions for switching to the prompt tab and using Generate Preview to assemble it, while Run Task is described as a future feature; the final caption suggests text appeared but does not establish that the full implementation was complete.
- 2:10:58
Optimizing Task-Specific Agent Prompts
The speaker tested a system that generates agent instructions after reviewing the codebase, history, and project documents. The first result appeared excessively long and generic, so he revised the instructions and compared previews across different tasks. He concluded that the context, task, and prompt need tighter specification, followed by further optimization and testing before relying on the result.
Got a question about something in this stream?
Ask in the group