Building AI Apps Live from Cairo 🌙
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 combines practical work and early-stage exploration around AI-assisted building and agent coordination: taking a site live after moving DNS to Cloudflare and discussing the expected GitHub-to-Cloudflare deployment flow, then expanding an Electron swimlane timeline into a unified command center for roadmaps, tasks, dependencies, and agent state. Luqman discusses a phased, partly parallel implementation using Markdown plans, shared JSON state, and a human-review queue, while deferring OpenClaw integration until the core tool is ready. Separate-project troubleshooting established that the Electron tool had been created apart from the main project, with a tracker described as the bridge between them; reference documents were intentionally left at their canonical paths to avoid diverging copies. The stated development stack includes Supabase, Railway, Cloudflare, and Codex. Other discussions cover dividing work between Claude and Codex, context-window consumption, Warp, local versus VPS operation, usage limits, and the API alternative. The stream also describes an automated clipping workflow, with possible later redistribution on platforms such as TikTok. Audience questions touch on Egyptian dialect support and coding tools, while broader topics include an English-first interface with selective Arabic cues and a vision for growing an Arabic vibe-coding community through transparent live building. Luqman argues, as an opinion, that easier building does not ensure shipping without marketing, distribution, and profitability knowledge. Several comparisons remain unresolved, and the command center remained an unfinished concept and prototype.
37 moments from the stream
- 0:51:41
Regularly Live-Streaming the Building Process
A participant suggests that regularly live-streaming the building process on YouTube could be a relatively open regional opportunity, arguing that vibe coding is accessible to people with limited backgrounds and that ongoing transparency may attract viewers interested in the work. It is presented as time-dependent advice, not a guaranteed outcome.
- 0:54:12
A Knowledge Hub for the Vibe-Coding Community
Luqman describes a knowledge hub for vibe coders, potentially focused on Arab learners because the North American content space is crowded. He distinguishes this educational aim from building an audience primarily to sell products to it.
- 0:57:26
Building Is Easy; Shipping Requires Business Understanding
Luqman argues that modern coding tools have made ideas easier to build, while many developers still fail to ship because they lack marketing, advertising, and distribution knowledge. He suggests drafting a manifesto and roadmap that address profitability and a route to market, while noting that some business models may not be profitable at all.
- 1:00:24
Designing a Project and Agent Command Center
Luqman explores an early concept for a desktop app combining a swimlane project map, task board, shared state, and agent hub. It would expose infrastructure, milestones, dependencies, and current work so agents could follow weekly execution, while the swimlane could be updated when delays occur. The material shown was conceptual and mocked up, not a completed product.
- 1:02:26
The Development Stack in Use
Luqman says his development stack includes Supabase, Railway, Cloudflare, and Codex. One additional tool name is omitted because its transcription is unclear.
- 1:09:00
An Automated Stream-Clipping Workflow
Describes setting up an automated system to clip streams, with a possible future plan to repost clips on platforms such as TikTok after enough live recordings are available.
- 1:12:36
Designing an Agent Hub as an Agent Command Center
The concept shifts from a swim-lane interface toward a command center, prompting a request for a practical explanation of the Agent Hub and examples of organizing OpenClaw agents hierarchically for planned independent work. Codex and Claude are mentioned as current tools, while the command center remains a pre-build concept.
- 1:16:37
A Shared Command Center for Coordinating AI Agents
A concept for a monitoring and coordination layer where multiple agents and tools read from and write to one shared JSON file, while an Electron interface watches changes and updates project state in real time. An orchestrator could delegate work to specialized agents that report results back to the tracker, with drafts and proposed actions placed in a human review and approval queue.
- 1:21:30
Deferring Integration Until the Core Tool Is Ready
The speaker judged that integrating OpenClaw into the command center before setting up the tool itself would be premature, so the idea would remain conceptual and the integration would be deferred to a later stage.
- 1:23:14
Phasing the Command Center Build for Parallel Work
Rather than building the system in one shot, the discussion proposes an executable Markdown plan grounded in reference documents such as a manifesto and roadmap. A mandatory foundation phase establishes the app scaffold and server; phases 2–4 can then proceed in parallel across agents, while a final phase covers policies, keyboard shortcuts, ripple effects, and verification. The foundation output should be validated by the user before proceeding.
- 1:30:18
Warp as an Agentic Terminal Rather Than a Full IDE
The speaker characterizes Warp as an agentic terminal rather than a full IDE and notes that its drag-and-drop experience can be somewhat awkward.
- 1:41:22
Executing a Multi-Phase Plan with Agents
A workflow for instructing an agent to read a Markdown file and execute a numbered task, completing the first phase before running several later phases in parallel and using Codex for a subsequent polishing phase.
- 1:43:40
Combining a Roadmap with an Agent Task Board
The concept evolved from two separate tools—an agent command center and a swimlane timeline for roadmap stages—into a unified task board with an agent hub. The mentioned shipping and optimization weeks were planning targets, not achieved results.
- 1:45:42
Troubleshooting a Desktop App Created as a Separate Project
While trying to view an update inside the Electron app rather than on localhost, the discussion identifies that phase one created the development tool as a separate project from the main project, with a tracker described as the bridge between them.
- 1:50:14
Avoiding Duplicate Project Documents and Configuration Drift
A discussion of which files belong in the separate command-center project. The recommendation was to leave the manifesto, roadmap, and submission checklist at their referenced paths to avoid diverging copies, and place only the new project's own build specification inside it.
- 1:53:28
Expanding a Swim Lane into a Command Center
Luqman explains that discovery led to expanding the work from a single swim lane into a command center, with deeper project details and its countdown to follow after that phase. The stated timeline was an estimate made during the stream, not a confirmed outcome.
- 1:54:32
A Vision for an Arabic AI Coding Community
Luqman describes the channel’s goal of building an Egyptian and regional community around AI-assisted coding by streaming the work transparently instead of only publishing short explanations of specifications and models. He believes practical building may encourage people without a web-development background, while voice interaction with agents and the stream archive help him review the project’s evolution.
Got a question about something in this stream?
Ask in the group