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
- 3:55:44
Codex drafts and the adversarial lens
Describes a Codex follow-up draft that can be sent with Tab. Explains its adversarial lens as a suggestion that challenges the design, while noting that the user may already be satisfied and prefer to stop iterating.
- 3:56:55
Brainstorming a “Talk to the hoopoe” invitation
Proposes a subtle invitation to interact with the hoopoe while keeping the main heading, drawing on the symbol’s familiarity to Arabic-speaking audiences. A teaser that moves away from the pointer is brainstormed without a confirmed implementation.
- 3:59:32
Open source needs a vision beyond a temporary feature
Discusses a request to open-source an agent orchestration platform and develop it through community contributions. The speaker argues that rapid AI tooling changes require a long-term vision, since major platforms may absorb capabilities previously served by standalone tools.
- 4:01:45
Making the English interface playful too
Returns to the English interface and asks how to give it a playful character too, expressing dissatisfaction with its current appearance. No final solution is specified.
- 4:02:15
Requesting trade-offs when design recommendations disagree
Notes conflicting system-design recommendations and asks for the recommendations and their trade-offs. The supplied pages end before an answer, and the parties’ names are not clear enough to identify reliably.
- 4:02:42
Expanding system design with another agent’s review
Luqman describes asking to expand an MVP-level system design into a fuller roadmap with another agent’s input. He requests combined recommendations, trade-offs, and research where needed to resolve disagreements.
- 4:05:25
Difficulty viewing multiple Codex sessions
Luqman criticizes the difficulty of viewing multiple Codex sessions simultaneously and looks for a way to display them separately. His tentative exploration does not establish whether the feature is available.
- 4:06:44
The future of SaaS compared with e-commerce
Luqman discusses a forecast attributed to Amjad Masad that software generation could sharply reduce SaaS valuations. He argues the shift is not immediate and advocates shipping products and observing user response now. He considers e-commerce scalable, especially with an owned brand, but views its margins as weak.
- 4:10:04
Reviewing agent recommendations and steering the work
Luqman reads a joint recommendation from two agents, then requests an overview of the finished application’s sections to clarify his goals. He describes shifting from detailed implementation instructions to intervening when the direction needs to change, while the agent carries the work forward.
- 4:12:21
Trust in AI forecasts and discussion of risks
Luqman links his trust in expert forecasts to developments he believes have materialized, referring to late 2025 and October 18 in his comparison. He then discusses a warning attributed to Geoffrey Hinton about the risks of intelligence surpassing humans, presenting it as a risk forecast.
- 4:14:18
Must you understand every line? Sales and distribution
Answering an audience question, Luqman argues that building a project does not require understanding every line or every technology, while recommending knowledge of the fundamentals. He emphasizes sales, distribution, partnerships, marketing, traffic, and visibility.
- 4:15:57
Micro-SaaS and focused solutions for businesses
Luqman presents micro-SaaS and business services as alternatives to a large, time-consuming project. He suggests identifying specific company problems and offering project-based solutions using AI.
- 4:16:54
Assessing Hormozi’s content for learning sales
Luqman praises Hormozi’s experience and books but considers his content more motivational and broad than detailed guidance for beginners. He believes it may be more useful to people with prior experience.
- 4:18:35
A view on freelance opportunities through Upwork
Luqman considers Upwork potentially slow but still a viable freelance channel, citing success stories he says he has encountered. He briefly mentions using AI without explaining a specific method on this page.
- 4:19:31
The value and limits of paid mentoring
Luqman discusses business owners seeking mentors when growth stalls. He cites an account of Justin Sun’s lunch with Warren Buffett, corrects his quoted amount to about $4.5 million, and suggests some expensive meetings may serve publicity more than reveal a secret to success.
- 4:25:19
An endorsement of a magnesium supplement
Luqman recommends a magnesium supplement and describes it as important and expensive, without explaining specific benefits or presenting evidence in this discussion. The precise formulation is unclear.
- 4:26:23
Active sessions prevent an app restart
Luqman says running sessions prevent him from restarting the app and proposes switching to Chrome temporarily. He also reports declining computer performance; these pages do not document a restart or a resolution.
- 4:26:52
How does a Jarvis-like assistant work?
Answering an audience question, Luqman explains interconnected agents coordinated by a central assistant. He tries requesting a briefer, more professional tone, calls the demonstration experimental, describes messaging and phone calls as reporting channels, and discusses adapting such systems for different users.
- 4:33:07
Concern about shrinking free disk space
Luqman comments on a low-disk-space warning and expresses surprise at the decline in available space over several hours. These pages contain no diagnosis or documented cleanup.
- 4:33:24
An opinion on Claude’s new interface layout
Luqman says he dislikes Claude’s new interface layout without explaining his reasons in detail.
Got a question about something in this stream?
Ask in the group