A Distraction Worth Following?
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 Luqman’s still-incomplete custom “Mission Control” platform for managing AI agents, tracing their work and usage, and reviewing their outputs. He practically explores a pipeline that turns stream recordings into transcripts and candidate clips, then supports human review, trimming, revision notes, and eventual scheduling; testing exposes problems with trimming logic, displayed duration, and interface clarity, so the work is not presented as a finished success. He explains that the return to custom development is motivated by control, traceability, and useful outcomes, with open-sourcing considered only if the system proves valuable. Audience questions broaden the discussion to SaaS marketing: in his view, AI may make building easier but cannot guarantee traffic, while market entry still requires distribution, differentiation, competitor research, and an early monetization plan, with different approaches for B2B and B2C. He also discusses bootstrapping versus funding, choosing tools and models by quality, speed, and usage constraints, the cost of excessive agent-to-agent conversations, and the uncertain future of education and work under AI without offering definitive predictions. Other themes include maintaining a clear vision, welcoming respectful disagreement in communities, structuring streams around measurable goals, and untested future ideas such as an Arabic voice assistant.
94 moments from the stream
- 3:20:07
Choosing a Website-Building Tool and Prioritizing Performance
Responding to a question about building websites for clients, the speaker says some options are close but expresses a preference for Codex based on overall efficiency and certain coding and media-generation tasks. In the practical demonstration, he describes the initial result as visually unpolished, while prioritizing performance at this stage and leaving visual refinement for later.
- 3:25:24
Debugging Clip-Trimming Logic and Building an Automatable Workflow
A hands-on test reveals that the trimming logic appears to do the opposite of what was intended, keeping the selected interval instead of removing it. The speaker also discusses review and approval, silence adjustments, direct export, and logging edits to reduce manual review gradually, while acknowledging the risk of full automation.
- 3:30:56
How to Start Using AI: Context Matters More Than Being a “Professional”
The speaker argues that AI skill is not determined by a “professional” label or hours of use alone, because people use it differently and with varying effectiveness. Rather than prescribing the same starter project to everyone, he now recommends talking naturally with a capable AI assistant, supplying relevant context about one’s background and goals, and asking for a suitable starting path, while noting that this advice could be perceived as asking people to surrender their own judgment.
- 3:34:16
AI for Goals Beyond Building Software
The speaker revisited his earlier, narrowly focused advice to use Codex for building, noting that people have different backgrounds and ambitions and can apply AI elsewhere, such as in their studies.
- 3:36:24
A Personal Assistant That Understands History and Goals: Benefits and Risks
The speaker discussed a tool that analyzes a user's history to understand their background and goals and provide personalized guidance, while warning that such comprehensive tracking could give the tool substantial influence over the user's decisions and behavior.
- 3:38:36
Simplifying a Complex Interface Through Clipping-Editor Research
The speaker criticized the interface as ambiguous and overloaded with buttons, asking for a simpler, more direct experience. He suggested researching existing clipping-editor examples, selecting the strongest approach, and then implementing it.
- 3:43:25
A Clean, Review-Friendly Clipping System
Luqman describes wanting a YouTube-like clipping system that produces clean clips and makes review easy, even if processing takes longer.
- 3:47:01
Do Newer Models Still Need Complex Prompts?
While examining a shared Markdown file, Luqman and the audience try to determine whether it presents a jailbreak or a reusable prompting system across models. Their tentative interpretation, still qualified by uncertainty, is that it discusses how newer models may need simpler guidance than traditional elaborate prompts and is said to include a leaked system prompt.
- 3:59:06
An Opinion on Funding vs. Bootstrapping
The speaker argues that bootstrapping can become harder at some stages of life and experience, that seeking an investor may be appropriate after initial success, and that specialized investment groups exist.
- 4:00:02
Testing Video Section Removal and Edit Delivery
The speaker tests removing multiple sections from a video and previewing the result, encountering a temporary glitch before a later attempt works. He then asks why the displayed video length does not decrease and how to finalize and deliver the edits, exploring closing the removal mode and then requesting changes.
- 4:03:43
Plan Monetization Before Building the Project
The speaker recommends defining a monetization strategy before starting to build a project instead of postponing it until after implementation.
- 4:07:46
Structuring Streams Around Measurable Goals
The speaker proposes a consistent stream format with an introduction, milestones, goals, and an agenda, emphasizing tangible, measurable outcomes followed by an end-of-stream review.
- 4:08:31
Reviewing the Development of an Agent Management System
The review covers adding a trimming section to the automation while noting that captions and silence handling still need refinement, plus workspaces for previewing and approving departmental output. A conversations section was also added for interacting with agents and viewing their files and department data, with possible agent-to-agent conversations envisioned later.
- 4:12:45
Choosing a Suitable Streaming Schedule
The speaker considers alternating streams between mornings and evenings and says midday streaming is not his preferred option.
Got a question about something in this stream?
Ask in the group