Back to Warp.dev? 2x Claude + Codex
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 indexed portion of the recording centers on coding-agent orchestration. Luqman describes using Codex to supervise Claude sessions through computer use and scheduled checks, allocate tasks, hand off context, and review work. Audience discussions raise workflow visibility, lost details, and trust in reviews, particularly for scientific code; reliability and token efficiency remain unresolved. He explores a goal-driven workspace combining agent teams, memory, roadmaps, and an interface that adapts to each project. Practical work includes reviewing system architecture, extending plans beyond MVP, and requesting landing-page and hoopoe-animation revisions. The concept and revisions remain in development, alongside discussion of open sourcing; completed implementation and fixes for computer-performance problems are not established. Audience questions cover getting started with AI, programming fundamentals, hardware requirements, security, and developing skills. Other discussions examine task cost versus token pricing and model speed without a measured comparison. On product strategy, Luqman emphasizes lasting usefulness, discovering problems for a focused audience, and planning marketing and distribution. He argues that easier building does not guarantee sales and that marketing outcomes remain uncertain. Secondary topics include agents that read and reply to Facebook and TikTok comments while identifying themselves as AI. Luqman reports workflow problems, local slowness, and agents stopping when the application or computer closes, without documenting fixes. Participants also discuss voluntary creator support through tips and paid messages, affiliate marketing, sponsorships, personal branding, and professional visibility through building in public; the question of how personal branding relates to technical competence remains unresolved. Discussion also covers AI-company IPOs and profit-and-loss disclosure. At recording time, Luqman offers unverified forecasts about IPO timing and funding improving pricing and efficiency, while audience contributions suggest funding could support data centers and research. He also discusses his trust in expert forecasts and a warning attributed to Geoffrey Hinton about superhuman-AI risks, presented as a risk forecast rather than an established outcome.
160 moments from the stream
- 4:34:06
A question about moving to Cairo while working remotely
An audience member asks whether moving to Cairo is worthwhile while working remotely and what comfortable living might cost. Luqman begins by clarifying the intended city; these pages provide no answer or cost estimate.
- 4:36:42
Does remote work require moving to Cairo?
The speaker argues that a remote worker need not move to Cairo if reliable internet is available where they live.
- 4:38:00
Orchestrator choice and computer-use capabilities
Asked about an alternative orchestrator, the speaker says he has not tried it but expects it could be capable. In his opinion, GPT or Codex has stronger computer-use capabilities than Claude; the specific model names are unclear in the captions.
- 4:43:46
Personal branding, technical competence and creator income
An audience question asks whether personal branding overshadows technical competence. The response focuses on visibility, affiliate marketing and sponsorships, with the speaker describing YouTube advertising returns in Egypt as limited; it does not settle the comparison with technical competence.
- 4:46:47
Building in public and professional opportunities on LinkedIn
The speaker discusses how sharing work while building it may improve professional visibility and hiring opportunities on LinkedIn. Based on his observations, he considers building in public more common outside Egypt.
- 4:47:39
Feedback on Claude’s updated session interface
The speaker likes the session rail in a Claude update but dislikes another session section. This is brief subjective feedback without a detailed explanation.
- 4:48:44
Delaying a GPT update because of cron jobs
The speaker says recurring cron jobs kept him from closing GPT to update it, then announces an intention to update. These pages do not establish that the update completed.
- 4:50:27
Requesting a quick main-area mockup
The speaker requests a simple, quick mockup of just the main area to help him understand it. These pages contain neither the resulting mockup nor enough context to identify the project.
- 4:52:33
Finding product ideas through everyday problems
Answering an audience question, Luqman suggests starting with problems in the market and everyday life, arguing that personal experience strengthens motivation to solve them. He also suggests giving AI explicit idea criteria, such as profit margins, fixed versus variable costs, and limited competition.
- 4:54:45
Building alone does not guarantee sales
Luqman argues that building a product alone is insufficient to sell it and emphasizes how challenging the market can be. The audience discussion highlights the importance of marketing as building solutions becomes easier.
- 4:59:28
AEO competition and the role of visibility
Luqman argues that similar offerings in the AEO market make visibility and funding decisive, with product quality alone insufficient to win. This is his assessment of the market, not an independently verified finding.
- 5:00:24
What determines readiness to start a startup?
Answering a question about finding a startup idea, Luqman ties advice to skills, experience, resources, runway, and investor connections. He says a useful recommendation requires understanding the person's circumstances and identifies marketing as a potential obstacle.
- 5:01:57
A focused audience and marketing before building
The discussion favors serving a small, specific audience to clarify targeting and marketing. Luqman also associates narrower targeting with higher cost per click, then encourages understanding the audience, market, and acquisition approach before building the product.
- 5:03:06
Planning all application sections in one layout
Luqman requests a single layout showing the application's categories and sections as envisioned when complete, while deferring visual revisions. This establishes a structural request rather than a completed implementation.
- 5:07:05
The limits of marketing guarantees
Luqman discusses the difficulty of guaranteeing sales growth or conversion rates, distinguishing statistical estimates from promises of certain outcomes. The discussion covers limits of AI advice and “results or you don't pay” offers while maintaining that market outcomes remain uncertain.
- 5:09:12
An adaptive workspace for goals and project detail
Luqman describes an early AI environment concept involving harness engineering, acknowledging difficulty articulating it. He requests project overviews and detailed views, roadmaps, goal tracking, and sections added automatically for project needs, such as analytics.
- 5:12:29
A workspace that adapts to user interests
Luqman imagines an AI workspace that adapts to user interests, such as creating an analytics section for someone interested in analytics. The term he is trying to recall remains unresolved.
- 5:13:12
Personalization value and limits of results-based services
The discussion considers the view that personalization could command higher prices, then turns to results as a service. Luqman suggests this model suits certain services because future outcomes cannot be guaranteed for every service.
- 5:17:29
A question about 2D characters for a game prototype
An audience member asks for help designing 2D characters for a game prototype. Luqman begins referring them to a community with people more experienced than himself; these pages contain no design method.
- 5:21:37
A goal-driven AI workspace concept
The speaker proposes a workspace for coding and running a business, with project management, roadmaps, missions, and tracking goals and ambitions built into its foundation. Skills and MCP integrations are proposed as supporting tools, alongside a possible manifesto to clarify the vision. This is a proposed direction, not a confirmed implementation.
Got a question about something in this stream?
Ask in the group