Astra Managing Two Claude Accounts
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
This indexed portion focuses on Luqman’s account of coordinating Claude sessions through Astra and Codex using requirements documents, roadmaps, periodic follow-ups, and draft prompts. He sees this as movement toward greater automation while retaining human direction. Tracking parallel work through Mission Control remains a recurring problem; the records do not confirm its repair or completion of project milestones. He also explores a workspace that adapts to user goals, outcome-based token estimates, and voice interaction, with feasibility, costs, and delivery timelines unresolved. Audience questions cover model selection, usage limits, assistant agreeableness versus forced criticism, and whether QA can keep pace with AI-assisted development. Luqman suggests AI-assisted testing under human supervision without presenting a practical evaluation. Other discussions emphasize marketing, distribution, and sufficient traffic to assess product performance, alongside his view of intense Shopify competition. Beginner guidance focuses on discussing goals with an assistant, learning through building, and sharing work publicly. Luqman also discusses his expectations for AI pricing and broader access at different price levels, including potentially improved efficiency and lower prices following IPOs. He speculates about model launches and IPO timing, including his recording-time prediction of a possible OpenAI IPO in Q1 or Q2 2027. He also imagines AI companies expanding into dedicated assistant hardware, such as necklaces and pocket devices; these are forecasts rather than confirmed policies or events. Creative work includes directing noir-jazz and Eastern-influenced focus music through Suno Studio, reviewing character movement and seamless visual loops, and exploring a hoopoe mascot and a bounded website explainer. Advertising interruptions, stream quality, and encoding load also receive attention. This portion combines explanations, exploration, and revision requests, with several work outcomes remaining unconfirmed.
165 moments from the stream
- 3:12:22
Start with platforms or a local agent?
Luqman recommends starting with platforms in response to a question about building a local agent, and says the value of a higher subscription tier depends on the project.
- 3:13:22
Estimating tokens by the desired outcome
He proposes estimating a token budget for a complete outcome, such as finishing a website, and getting agreement on that budget rather than focusing on individual tasks.
- 3:14:52
Managing Claude sessions through Astra/Codex
He explains discussing the project vision and preparing a roadmap and requirements documents before asking Astra/Codex to supervise execution in Claude’s interface through computer use. He describes monitoring several sessions in parallel, while acknowledging that some viewers encountered difficulties whose cause he does not know.
- 3:16:52
Questioning development timelines for continuously running agents
Luqman challenges model estimates spanning weeks or a month, arguing that they do not account for continuously running AI agents in the 2026 context he describes. He requests a revised estimate and offers a shorter expectation without presenting a measured completion time.
- 3:20:06
A question about Antigravity and a Gemini model
Asked about Antigravity and a Gemini model whose name is unclear in the captions, Luqman says he has not tried Antigravity and has limited experience with the model, so he offers no firm assessment.
- 3:23:21
Naming music to convey its genre and oriental touch
Requests naming options that convey the music genre and an oriental touch, noting that the current name omits the genre. No final choice appears in this passage.
- 3:25:25
Open source or subscription: the role of operating costs
An audience question prompts discussion of open source and subscriptions, with the decision tied to service delivery costs, margins, and sustainable development. The model remains undecided.
- 3:27:29
Exploring voice integration with Whisper and ElevenLabs
Requests a review of the voice integration stack and the OpenAI Whisper repository, including its capabilities and whether ElevenLabs will be used. This is a research request, with no implementation or testing established.
- 3:28:59
A limited free trial to gather user feedback
Proposes a free trial with limited duration and registrations to gather substantial user feedback. The participant group size remains undecided.
- 3:29:35
Managing Claude sessions through Astra and Mission Control
Describes using Astra and Codex to coordinate Claude sessions and distribute work through requirements documents, missions, milestones, and a roadmap. He argues that a central Mission Control helps track parallel work as sessions become difficult to follow, alongside brief human direction and token usage monitoring.
- 3:34:54
Defining voice interaction requirements and stack priorities
Luqman requests a primarily spoken experience with near-real-time interaction. He explains that stack recommendations depend on the priorities and constraints stated in the prompt; these are proposed requirements, not implementation results.
- 3:37:05
Retracting an estimate does not settle the timeline
An audience question points out that the implementation timeline remains unresolved. Luqman explains that the earlier response retracted its statement without confirming that the work could be completed in a shorter period.
- 3:39:42
Traffic as a prerequisite for measurement and Shopify competition
Luqman explains that assessing pricing and user experience requires traffic and data, including drop-off and bounce metrics. He expresses the view that competition within Shopify’s closed ecosystem makes traffic acquisition difficult and advises against entering it in this discussion.
- 3:41:53
Merchant visibility in AI answers and its relationship to search
An audience member asks how AI answers relate to search engines and whether SEO is needed. Luqman connects the question to a Shopify app intended to improve merchant visibility in AI answers, without presenting evidence of its effectiveness here.
- 3:43:09
Describing Codex oversight of a Claude session and code
Luqman describes using Codex to follow activity inside a Claude session and oversee the code as well. The captions provide his account of the workflow without independently establishing its capabilities or outcomes.
- 3:48:18
Trying LinkedIn for product awareness
An audience member asks about discussing the product on LinkedIn. Luqman says he is trying that approach, without reporting results.
- 3:48:40
Late coffee and sleep quality
Luqman discusses how caffeine remaining in the body may affect sleep quality even when someone can fall asleep. He contrasts this with anecdotal claims that coffee helps some people sleep; the discussion is informal, not a test or medical assessment.
- 3:52:13
Advanced voice experience costs and implementation priorities
Discusses distinguishing features worth investing in, then asks about the cost of an advanced voice experience and ways to reduce it. The scope is to be agreed before adding it to the current milestone; the page establishes no firm cost or timeline.
- 3:57:42
Investigating encoding overload during streaming
Requests an investigation into encoding overload and high CPU load while streaming and using Codex and Claude. The speaker suspects that uploading a YouTube video during the stream may contribute, but the page confirms neither the cause nor a fix.
- 4:05:22
Stream lag during background 4K video processing
Luqman questions whether a Codex session is still processing a video in the background and needs to be stopped. He later attributes the stream lag to CPU and encoding load during 4K video processing and reports that the problem is fixed, without explaining the remedy. The discussion also touches on device memory and capacity.
Got a question about something in this stream?
Ask in the group