New Project? New CC?
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 centers on a vision for Mission Control as a unified layer for monitoring agents and sub-agents, routing work by goals and available capacity, and moving from small tasks toward longer, more autonomous builds. Luqman discusses preparing PRDs, ERDs, system design, infrastructure, and access in advance while keeping humans in the loop and adding code, security, and Playwright checks before merges; however, the design, integrations, and full automation remained exploratory, with no confirmed completion. Audience questions covered choosing tools such as Codex and Claude, context and usage management, learning Python versus AI engineering, and reverse engineering with its legal risks. Practical discussions also addressed product distribution and audience reach, email-deliverability testing, AEO for Shopify stores, interface improvement, and repository knowledge. A recurring attributed view is that focus, quality, distribution, and reviewing AI-produced work matter more than maximizing project or content volume, while several technical, legal, and experimental outcomes remained unresolved.
103 moments from the stream
- 0:06:29
Technical planning foundations before building a project
The speaker describes starting a project with a PRD, an ERD, and system design, while also considering security and Playwright end-to-end smoke testing.
- 0:07:49
Mission Control versus graph engineering
The speaker distinguishes graph engineering for ongoing operations in an established organization from Mission Control, a simpler view for tracking agents, tasks, and the progress of a new product build. He also discusses customizing an existing tool and possibly releasing a general open-source version later.
- 0:12:12
Quality over quantity in social media content
The speaker shares advice he agrees with: publishing fewer high-quality pieces may be better than producing many low-quality social posts.
- 0:14:01
Mission Control as a map for ideas and work
The speaker compares a command center to a map for organizing ideas and tracking progress toward a goal, potentially through visual workflow lanes. He regards Jira and open-source alternatives as valid options but personally prefers a custom tool when control and customization matter.
- 0:16:19
Monitoring agents and long-running coding sessions
The speaker presents an early interface concept that shows agents and sub-agents working on tasks in one place. Its purpose is to track progress and understand what happens during long-running coding sessions, while exploring a unified view of Codex and Claude work.
- 0:20:47
Balancing Sub-Agents and Code Audit Crews
A qualified discussion of whether multiple agents use more tokens in exchange for finishing faster, followed by a proposal to run an audit crew before merging a pull request and review the code through lenses such as security, coherence, and compliance.
- 0:23:00
PRD, ERD, TDD, and End-to-End Testing
A practical distinction between product requirements and entity-relationship documentation, with the observation that an ERD may need to be requested explicitly from an agent. The discussion then turns to test-driven development and mentions Playwright for end-to-end testing, while leaving its exact place in the TDD workflow unresolved.
- 0:24:07
Starting with One Focused Project
Advice for a newcomer with several application ideas to begin with one web project, develop it deeply, and avoid spreading effort across more work than current capacity or usage limits support.
- 0:25:39
Simplifying a Multi-Project Mission Control Around One Orchestrator
A redesign brainstorm for an outdated mission-control interface: a central orchestrator that launches tools and sub-agents only when needed, a clean project overview, pull-request tracking, familiar left-side navigation, and an explicit way to switch among multiple projects. A clearer, more lively visual representation of agents was also proposed.
- 0:30:51
An AI-Controlled Kanban Board for Project Management
An explanation of a product idea combining a Kanban board and project-management software with AI-driven control, inspired by tools that distribute and track work across teams.
- 0:32:22
Exploring AI-Assisted Reverse Engineering
An exploratory discussion prompted by a question about using AI for game reverse engineering. A participant offered a simplified, unverified account of analyzing binaries to recover code logic; the speaker then considered whether the idea applies beyond games and what it means for protecting distributed files. The discussion explicitly notes that reproducing proprietary software this way may be illegal and is neither easy nor guaranteed.
- 0:37:31
When Should Legal Steps Enter a Product Timeline?
An unanswered general planning question about whether legal steps, such as engaging a lawyer, are needed before or after building an application, and the need for a clear timeline for that work.
- 0:38:14
What Are Binary Files?
A brief explanation connects machine language with binary and describes binaries as the files that run an application on a machine.
- 0:39:33
Accessing App Logic in Open vs. Closed Environments
A question asks whether inspecting application logic differs between a PC app and a closed environment such as Shopify; the page does not record an answer.
- 0:39:58
Tracking Agent Progress with JSON and Goals
A discussion of using a JSON file to give agents project status, possibly cleaning it up if it grows too large, and defining a clear goal they can follow.
- 0:40:56
Video Quality Over Quantity
A practical opinion that publishing fewer, higher-quality videos may be better than continually increasing volume without clear results.
- 0:41:26
Choosing a Plan and Optimizing Codex Usage
The speaker says plan differences mainly concern usage limits and some tools rather than intelligence, suggesting that users choose based on whether they reach the limit. For optimization, he recommends adjusting effort by task and cautions that some optimization skills may reduce output quality if used without understanding them.
- 0:45:07
Switching Effort vs. Switching Models
The speaker tentatively explains that changing effort within a conversation does not reset its context, while switching models may sometimes behave differently in how the conversation is reread or consumed.
- 0:46:43
Safely Diagnosing Storage Pressure During Streaming
A discussion of low storage that can interrupt recording during a stream. The proposed approach is to request analysis and a plan first, then inspect the exact action before allowing the tool to perform it.
- 0:49:57
Migrating a Shorts Batch to an Updated Workflow
The review notes that the October 7 shorts batch still uses an older version and proposes updating it and trying the new workflow on that batch.
Got a question about something in this stream?
Ask in the group