Inbound vs Outbound Traffic
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 widening gap between building products with AI—which is becoming easier—and the harder work of distribution, customer acquisition, and sales. Luqman reassesses a heavy reliance on inbound content and considers a broader mix of outbound outreach, cold email, advertising, partnerships, and SEO, with the practical principle of testing channels and concentrating on those that show real conversion signals. Practical work includes reconsidering Burnish’s pricing by replacing a time-limited trial with a free plan, investigating a reported HTTP 500 error and an observability gap, and simplifying AI-agent workflow diagrams; the records do not establish that these efforts were completed or successful. The discussion also explores multi-agent systems organized around shared goals, context, memory, and explicit follow-up, with a preference for starting small instead of introducing early complexity. In comparisons of Codex, Claude, and AI providers more broadly, Luqman emphasizes balancing quality, speed, price, usage limits, and completed workload, while repeatedly qualifying his views as personal and treating mixed reviews and benchmarks as insufficient for declaring an absolute winner. Audience questions broaden the stream to SaaS development, AI-engineering portfolios, content automation and scheduling, visibility in AI-generated search answers, team size, personal branding, transparent affiliate marketing, and AI support for study or business management. Major unresolved questions include the best acquisition channel, the commercial value of content, the right agent architecture, and the changing choice between tools and subscription plans.
133 moments from the stream
- 2:26:51
Codex and Claude Speed on a Long Task
An audience member observes that Codex was very slow compared with Claude on the work shown, and the speaker agrees that Claude is somewhat faster while noting that the task was large. This is an observation from a specific experience, not a benchmark.
- 2:27:29
Solo Founder With AI or a Small Team?
The speaker discusses how working alone with AI can reduce cost and responsibility compared with hiring a team, without guaranteeing higher quality. In his comparison, a small team whose members use AI well will generally be faster and better than one person, while he also offers the opinion that AI has reduced his former need for a partner to exchange ideas with.
- 2:29:58
A Personal Codex–Claude Comparison at the Time of Recording
Drawing on his use of both tools, the speaker gives a personal assessment at the time of recording: he considers Codex more integrated and stronger for some agentic, visual-generation, and design work, while preferring Claude for architecture and logic and sometimes finding it more dependable. He frames the practical choice as one of efficiency and acknowledges both positive and negative experiences with each.
- 2:32:41
Comparing a Personal Assistant with Grok
The response says both assistants can reason with and help the user, while a personal assistant can also follow ongoing work and help carry it forward. It qualifies that answer quality depends on the question and no assistant wins at everything.
- 2:35:51
One Assistant with Parallel Tasks or Multiple Independent Agents?
The discussion distinguishes one assistant running separate tasks in parallel from multiple independent agents, then compares it with a cron-triggered session. The proposed distinction is continuity of context, memory, follow-up, and failure handling, while permissions, scheduling, and explicit follow-up goals remain necessary; a well-built orchestration layer can make the functional gap quite small.
- 2:40:02
Recurring Research and Morning Reports for a Personal Assistant
The speaker questions how much a persistent assistant is needed and proposes a bounded practical use: recurring research on selected topics followed by a morning report, rather than assuming the assistant will build a product or perform every function on its own.
- 2:46:36
Evaluating Codex and Claude by Efficiency and Cost
The discussion argues that choosing a coding assistant should consider work efficiency and token efficiency alongside input and output pricing. Because the cited figures and migration claims are attributed to the speaker and an unverified source, the useful takeaway is a comparison framework rather than a proven verdict on either tool.
- 2:52:52
How to Choose Between AI Models in Practice
A discussion of balancing output quality, completion speed, cost, and usage limits when choosing an AI model for coding. The speakers treat benchmarks as partial indicators and real-world experience as important but subjective, suggesting that task-level efficiency—or using more than one model—may matter more than declaring an absolute winner.
- 3:01:18
Prompting a Disk-Space Audit
A concise request for a comprehensive disk-space audit that reports current free space, identifies what consumes storage, and investigates why the disk fills up quickly.
- 3:02:36
Comparing Codex and Claude on the Same Task
He proposes giving both tools the same task to judge which is more suitable, saying that in his experience the difference is not large while Claude is more expensive. He then frames a research task comparing pricing-plan and token-use efficiency in light of Dev Day 2026 announcements, noting that the question requires substantial calculation and may not yield a definitive answer.
- 3:08:07
The Practical Impact of an Ultra-Fast Mode
He praises an ultra-fast feature while noting that its actual speed was not clearly presented. He calculates an estimate of roughly 13× and illustrates the possible impact with hypothetical examples: reducing a 30-minute task to about two minutes and a 90-minute task to about seven. These are estimates, not documented benchmark results.
- 3:12:16
Paid Acceleration and the Cost of Faster Execution
The speaker describes a fast mode in an unspecified paid plan that is said to complete work much faster while consuming more resources, and argues that access to capital can accelerate building. The tool’s identity and the precise speed claim are not independently clear from the captions.
- 3:12:46
Claude vs. Codex: Efficiency Beyond Tool Loyalty
The speaker discusses strong user loyalty toward Claude or Codex and says that, in his view, the distinction has become more about usage efficiency than overall output quality. He considers Codex more token-efficient, while remaining uncertain because expected plan changes may alter that calculation.
- 3:19:31
When B2B Needs Outbound Traffic
Contrasting his B2C experience with early-stage B2B needs, the speaker argues that B2B may initially require more direct outreach and outbound acquisition. He describes inbound as attracting an audience like a magnet, while outbound means actively finding and bringing in that audience.
- 3:20:37
Transparency in Affiliate Marketing
While discussing affiliate promotion for a service such as Shopify, the speaker stresses clearly telling the audience when a link earns a commission and criticizes content that praises a product primarily to generate referral income.
- 3:22:14
Evaluating Affiliate Programs Before Promotion
The speaker discusses referral and affiliate programs from Shopify and Hostinger, emphasizing that their value depends heavily on audience size and fit, while eligibility and payout timing may not always be clear. He suggests checking the partner portal and its referral and lead details rather than relying on broad impressions.
- 3:28:14
Financial Incentives Can Affect the Credibility of Content Recommendations
The speaker argues that some creators may overpraise services because of referral commissions, so financial incentives should be considered when evaluating recommendations. The discussion also notes that consuming too many such videos can waste time and create distraction, and suggests independently researching alternatives.
- 3:34:31
Prioritizing Outreach Over Lower-Impact Work
The speaker says the issue was not opposition to outreach, but becoming occupied with work that mattered less, while emphasizing that outreach is highly important.
- 3:34:55
How Sponsorship and Monetization Can Affect Creator Credibility
The speaker argues that recommendations grounded in experience deserve more trust, while sponsorships, affiliate incentives, and a growing audience may tempt some creators to exaggerate or favor revenue over credibility. He frames this as a general risk rather than a definite judgment about any individual and says products should be explained without overpromising.
- 3:39:29
Background Music That Supports Focus
The speaker describes preferring background music that fades into the work instead of demanding attention, associating that quality with better focus, before noticing that the audio mix had made it too loud.
Got a question about something in this stream?
Ask in the group