What Happens If This Works?
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 explores an early open-source tool concept that starts with user goals and coordinates agent teams to build and operate projects, potentially through voice. Discussion covers agent harnesses, memory, knowledge and context graphs, RAG, and custom development versus learning from existing tools such as Hermes. Feasibility and the eventual interface remain unresolved. Practical work includes repeated reviews of a bilingual teaser page, clearer messaging, bird animation, and audience polls to choose a name, without a final naming decision. Domains are discussed separately, with availability checks reflecting the time of recording. Blogging and SEO agent work includes requests to review articles while keeping daily publishing paused. A comment-reply system is described as retaining human approval. The discussions do not establish completion of these improvements. Audience questions examine token limits versus quality and when agents should retry or seek help. The discussions also examine Luqman’s requests for clearer roadmaps and verification before approval, practical evaluation of Codex and Claude, model-intermediary privacy, sustained use as demand validation before monetization, branding, marketing, and web-service pricing. Luqman reports that a Shopify app competitor was acquired within the two weeks preceding the stated recording date, October 16, 2026. He interprets this as a possible demand signal; no confirmed acquisition price is provided. The discussion includes a simplified explanation of IPOs and financial disclosure, using a hypothetical Anthropic listing as an example. Predictions about future interfaces and AI pricing after IPOs are explicitly uncertain, while some privacy and performance questions remain unverified. Luqman encourages proactive self-directed learning and asking questions, mentioning courses as a structured option without recommending a specific platform. An audience contribution describes a site being polished that collects tech and AI news and features skills, tools, and open-source apps. Discussion touches on making the project open source and monetization, without enough clarity to establish a revenue model. The discussion also covers disk space repeatedly filling after cleanup, keeping local recordings for higher-quality short clips, and ongoing background work on livestream transcription. No storage diagnosis or fix is established, and the transcription implementation is unexplained.
115 moments from the stream
- 0:08:21
A competitor acquisition as a Shopify market demand signal
Luqman reports that a Shopify app competitor was acquired within the two weeks preceding the stated recording date, October 16, 2026. He interprets this as a demand signal; no confirmed acquisition price is provided.
- 0:09:22
Proposing a worker for daily blog content and product visibility
Luqman identifies visibility and traffic as priorities and proposes adding a worker to the workflow to produce daily posts on the product website. These captions describe a plan, without showing implementation or results.
- 0:10:42
Work decomposition and token availability as a delivery constraint
He describes organizing work into milestones, missions and tasks, and argues that token availability constrains delivery speed on large projects. Faster completion with unlimited resources is hypothetical, rather than a tested result.
- 0:11:29
Backing and distribution as project-selection criteria
Luqman presents backing, funding or a clear distribution partnership as criteria for choosing projects to join. This is his stated preference, rather than an established universal rule.
- 0:13:00
A vision of AI that starts with a goal
Luqman sketches a future where users state goals and AI assembles teams to achieve them, reducing reliance on current interfaces and prompts. He presents this as technically unresolved thinking, with a proposed term rather than an established standard.
- 0:15:42
Reviewing an AI-generated website’s code
An audience question asks how to review the quality and security of an AI-generated website. Luqman recommends Codex based on what he has heard, discussing tangled-code cleanup, project size, and tool consumption. The follow-up clarifies that it is a regular website for SEO work; no review results are shown.
- 0:18:13
Testing demand through repeated use
Luqman considers an open-source release as a way to test a tool’s usefulness in real workflows. He proposes observing sustained community use before deciding whether to enter the market; the approach remains tentative.
- 0:19:09
Differentiating by combining existing capabilities
Luqman argues that finding an idea with no competitors is difficult and that differentiation may come from combining capabilities spread across several tools. He limits his novelty claim to his own research and stresses that the idea is still preliminary.
- 0:21:13
Clarifying a page headline with a visual explanation
While reviewing a teaser page, Luqman calls its H1 ambiguous and requests alternatives before changes. He proposes a visual comparison between earlier AI use and a future vision centered on stating goals; the captions end before the explanation is complete.
- 0:22:22
Goal engineering: from requesting tasks to delegating outcomes
Luqman discusses agents that translate goals into roadmaps, requirements documents and system designs, eventually working toward improving a specified metric. He speculates that graphical and command-line interfaces could become unnecessary, while acknowledging that his vision may be wrong.
- 0:25:34
Long sessions and the tradeoff between token limits and quality
He mentions a long session involving repeated attempts, then addresses limiting token consumption during autonomous work. He says limits are possible but may affect quality, treats them as part of agent harness engineering, and recommends inspecting skills that impose such constraints.
- 0:27:01
A command center interface that adapts to its purpose
The discussion considers an audience proposal to embed a terminal in a command center and adapt the interface to the task. The example is omitting the terminal tab for social media agents when they do not need it.
- 0:27:30
Speculation on AI pricing after public offerings
Asked whether prices will rise or fall, Luqman predicts that IPO funding could expand data centers, lower prices and improve efficiency. He proposes a scenario in which Anthropic moves before OpenAI and gains subscribers, mentioning 2027 for an OpenAI IPO. These are predictions made in the recording, and he explicitly expresses uncertainty.
- 0:32:16
Voice and video AI assistants
The speaker predicts that interacting with AI virtual assistants through voice and video will be part of the technology’s future.
- 0:32:58
What is an IPO?
A simplified explanation of offering a private company’s shares to the public, followed by a discussion of financial disclosure using a hypothetical Anthropic listing.
- 0:34:36
Configuring an AI employee for SEO blogging
Requests an AI employee to write and publish daily website articles, with a contract, system prompt, and place in the org chart. Also describes ending another project-based role when its assignment finishes.
- 0:36:20
Reviewing agent replies before automation
Describes an agent replying to Instagram and TikTok comments with manual review and approval, with automation considered later if its performance proves satisfactory.
- 0:37:13
Choosing a name with community input
Branches the chat to generate five candidate names for a community vote, then suggests that including domains could make the choice confusing.
- 0:38:56
An opinion on Anthropic pricing and usage limits
An audience member suggests that lower Anthropic prices and more usage-limit resets could threaten OpenAI’s position. The speaker agrees, then remarks on the complexity of an IPO decision.
- 0:40:53
Learning through initiative and questions
Encourages proactive self-directed learning and asking questions, while mentioning courses as a structured option without recommending a specific platform.
Got a question about something in this stream?
Ask in the group