Four Projects. One Flow.
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
Luqman and the audience explore a platform that would turn conversational or spoken goals into teams of agents for building products and running operations. Topics include roadmaps, role contracts, architectural oversight, and knowledge and context graphs; the concept remains under development and its feasibility unresolved. Token consumption, repeated file reading, review loops, and blocked parallel work recur throughout, alongside proposals for event-triggered supervision without a practical comparison establishing its advantage. Practical work includes reviewing an animated hoopoe website, exploring 3D and reverse-engineering tools, and demonstrating keyword search in a partially indexed stream library, with semantic search proposed. Content automation, reply approvals, and scheduling are also reviewed, including posting that reportedly continued after a pause request and unresolved performance and storage problems. Audience questions cover learning AI, choosing a specialization, and reaching customers; Luqman recommends planning distribution and using direct outreach for B2B. His predictions about voice interfaces and increasingly autonomous agents remain forecasts, while the work presented does not establish a completed platform or successful end-to-end automation.
153 moments from the stream
- 3:05:03
A searchable stream archive and local storage challenges
Luqman describes downloading stream recordings for transcription and searching for moments mentioning topics such as Codex. He links storage pressure to recording files and mentions deleting them promptly, while leaving a suggested cache contribution unverified.
- 3:08:42
Learning AI by building tools you need
Luqman advises a beginner to avoid rushing, follow AI developments, read regularly and build tools for personal needs or a micro-SaaS project using tools such as Codex if feasible. He predicts broad changes by 2031 while acknowledging that he cannot specify what will happen.
- 3:10:13
Reviewing visual quality and asking about limitations
Luqman flags a result he considers lower quality than the original, including details such as eyelashes, and requests a comparison and an explanation of limitations. This passage identifies neither the tool nor a confirmed cause or fix.
- 3:12:21
Planning customer reach before building a product
In the opening of an answer about reaching an audience, Luqman mentions advertising and distribution partnerships as ways to gain visibility and recommends developing a strong strategy before building. The page ends before the discussion is complete.
- 3:13:01
B2B marketing: visibility and direct outreach
Luqman argues that a good idea still needs visibility and that word of mouth needs an initial push. For B2B, he recommends prioritizing direct outreach and outbound marketing, citing his experience of relying too heavily on inbound.
- 3:15:54
Validate the product before seeking seed funding
Luqman recommends proving the product through a pilot and sales to a few companies before approaching seed investors. He highlights valuation when considering the equity exchanged for funding, while acknowledging that he does not know how investment groups operate in Egypt.
- 3:19:04
Studying AI engineering and choosing a university
Luqman encourages studying toward AI engineering but explicitly says he does not know how Egyptian universities teach the subject. He expresses a critical personal view of local education and suggests considering study abroad if feasible; these are opinions rather than a documented assessment of programs.
- 3:20:11
A brief discussion of payment service experience
Answering a payment service question, Luqman mentions experience with Stripe, PayPal, and Shopify Pay. He revises his recollection of Amazon Pay to note limited prior experience, without offering a detailed comparison.
- 3:23:29
AI education quality and choosing what to study
The speaker offers a personal critique of AI teaching quality in Egypt while valuing substantive formal education and arguing that study choices depend on the institution. He questions treating AI as a single discipline, pointing to its many areas of specialization.
- 3:31:58
Are these AI engineer responsibilities?
A brief reading lists software development from specifications, scalable and reusable logic, and assistance with task estimation. The speaker asks whether these are typical AI engineer responsibilities; the page provides no clear resolution.
- 3:34:58
Clarifying requirements for a bird that follows the user journey
The speaker challenges changing implementation recommendations and specifies a lively, cinematic bird that jumps, follows the website journey, and disappears and returns. Proceeding with the proposed alternative is conditional on meeting those requirements; implementation success is not established here.
- 3:35:50
An expectation of growing demand for AI engineers
The speaker describes the referenced job description as typical for an AI engineer and expects more companies to need the role in the future. This is an opinion expressed in discussion, rather than a labor-market finding.
- 3:38:16
A proposed platform combining automation management and automated building
The speaker proposes combining automation management and automated building in one platform, seeking a workflow that moves somewhat beyond graphical interfaces and command lines. Naming, details, and feasibility remain unresolved; development may depend on community participation, with early contributions and a possible competition discussed.
- 3:41:06
Encouragement to build projects during university
The speaker encourages students to explore AI and build a project or business while at university, suggesting that it could create income opportunities. The advice is encouragement, without a guaranteed outcome.
- 3:41:49
An AI employee catalog organized around business needs
An audience suggestion about business integration leads into a proposed browsable catalog of AI employees for roles such as social media posting and stream editing. The concept includes defined contracts and custom employee creation, distinguished from the automated-building side; these are proposed capabilities, not demonstrated implementation.
- 3:44:01
Managing building agents through goals
Luqman sketches an early concept for a lead agent that discusses user goals and coordinates building agents, with requirements documents generated within the workflow. He describes the proposed role as closer to a business partner and distinguishes product building from ongoing operations.
- 3:46:16
How might development environments change as agents become more autonomous?
Luqman discusses a shift from traditional IDEs toward agentic development environments. He speculates that terminals and detailed prompting could give way to requesting goals completed over hours, days, or weeks. He connects this vision to harness engineering and forecasts he partly attributes to Sam Altman, presenting 2027 as a cited possibility rather than a confirmed timeline.
- 3:48:50
The human role in reviewing AI results
An audience member argues that humans will remain necessary in the workflow, especially for details. Luqman agrees that the role depends on the domain and predicts a shift toward reviewing results, while expressing concern about managers trusting AI more than employees.
- 3:50:09
Studying business and marketing while learning AI independently
Asked about study choices, Luqman considers business or marketing studies combined with independent AI learning a reasonable path. He adds that specialized AI studies may suit someone aiming to become an AI engineer, presenting this as his view on the tradeoff.
- 3:54:12
AI visuals and the need for specialized 3D tools
Luqman considers Gemini and OpenAI stronger than Claude for visuals and believes 3D work requires specialized external tools. He says he has not tried the new model being discussed; his lukewarm impression comes from reviews.
Got a question about something in this stream?
Ask in the group