Debugging Core Product Engine
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 moving Burnish toward a Shopify-ready V1: reviewing its product engine and value proposition, auditing what is actually wired between the frontend and backend, and refining onboarding plus the Pulse, Mentions, and settings sections around Polaris and Built for Shopify expectations. Luqman discusses making background-agent work visible through a continuous activity feed and connecting research, action, and measurement in a closed loop, alongside demo-state and free-audit funnel ideas; whether these ideas are sufficient, and how attribution and measurement should work, remain unresolved. Practical work includes tracing intermittent HTTP 500 errors and missing CSS registration on the Pulse route, noting an initial Supabase connection problem and caution around resetting onboarding state in the database without establishing a diagnosis, and auditing branches, Git worktrees, and pull requests after mismatched development environments made preserved changes appear lost. Audience questions and side discussions cover context-dependent Shopify app selection, SEO/GEO and product metadata, browser agents and Codex/Claude workflows, broader Git and agent-orchestration guidance, evaluating agent skills, AI-assisted study, and the tradeoff between livestream latency and quality. Other exploratory discussions address Product Hunt’s suitability for Shopify apps, traffic measurement, advertising, and possible discrimination around product launches, as well as practical comparisons of AI-generated storefronts; some model names and versions are inconsistent in the captions, and perceived speed was not formally measured. The stated direction is to finish Burnish’s core, simplify onboarding, and conduct a thorough Shopify compliance review before expanding marketing, while acknowledging that limited development-store testing, expected V1 bugs, migration and performance issues, and several measurement questions still require verification and concrete resolution.
121 moments from the stream
- 2:37:26
Separating V1 Scope from V2 Tasks in a GTM Discussion
In response to a go-to-market phase question, the discussion says the displayed progress indicator does not tell the whole story: V1 is considered roughly complete, while most remaining tasks belong to V2. This is the speaker’s assessment, not a documented test result.
- 2:38:41
AI Audit as a Hook Product
The go-to-market strategy shifted to a landing page offering a quick, valuable audit, followed by an optional deeper report. The stated aim is to gauge merchant intent and the likelihood of a later install or subscription; the approach was still untested and planned for after the Shopify submission.
- 2:41:03
Interface Polish and Application Activity Card
The work includes centering a button that explains how measurement works, then moving to a Burnish application section using a card in Pulse to show application activity. Some implementation wording between these points is unclear in the captions.
- 2:42:10
Go-to-Market Measurement Dashboard
A dashboard in the go-to-market section brings together metrics such as spend, scans started and completed, qualified leads, traffic sources, and GMV. The explanation notes that it needs revision after the earlier cohort strategy was dropped.
- 2:43:49
Using a Free Audit to Lead Into a Paid Offer
A high-quality free audit is presented as a way to build trust and demonstrate value before asking users to pay for the full audit or an introductory paid offer.
- 2:44:31
Auditing Git Branches and Worktrees Before Merging
The discussion calls for reviewing completed work, identifying the active worktree and branch, checking what is useful or incorrect, and creating checkpoints before safely merging into main.
- 2:56:47
Sharing Shopify MCP Configuration Across Coding Agents
The discussion considers keeping the Shopify MCP JSON configuration in a shared or global location so Codex and other agents can inherit it, alongside pre-PR checks and an inventory of active branches and worktrees.
- 2:58:27
Finishing the Product Core Before Marketing
The plan is to prioritize the product engine completely, postponing website and marketing-funnel work until the app is ready and submitted to Shopify.
- 3:01:03
Restoring Demo Mode and Cleaning Up the Git Workspace
The work involves locating and enabling the existing demo configuration, then cleaning up branches and worktrees and merging what can safely go into main before development resumes, while acknowledging that worktree management has become complicated.
- 3:04:14
From V1 to Shopify Compliance and Lead Acquisition
The plan is to focus exclusively on the app until V1 is complete, then work through Shopify’s compliance checklist and policies. Shopify Dev MCP is said to have eased design and compliance work, while the top-of-funnel approach is being reconsidered around a lead-generating hook product rather than the earlier concept.
- 3:06:53
Guiding an Agent to the Pulse Demo Configuration
A coding agent could not locate a previously created demo configuration for the Pulse page, prompting a request for its exact location and clearer directions for the agent.
- 3:07:17
Is Product Hunt Suitable for Shopify Apps?
An exploratory question asks whether Shopify apps appear on Product Hunt and whether the platform suits them; the stated view is explicitly uncertain and based on an impression that Product Hunt may be stronger for B2C products.
- 3:11:41
Measuring Traffic from Product Hunt
A question about listing a product on Product Hunt and whether traffic arriving through the platform can be measured using its chart or analytics.
- 3:13:53
Audience Reach and Ads on Product Hunt
The speaker says Muslim users browse Product Hunt for products, then notices a promoted listing and asks whether ads can be run on the platform.
- 3:14:29
Reviewing a Squash-and-Merge Decision
After pushing the branch, opening the pull request, and seeing no GitHub conflicts, a squash merge is performed and the speaker asks whether that choice caused any loss. The supplied captions do not provide a final answer.
- 3:15:59
Discussion of Discrimination and Product Launches
An inconclusive opinion discussion considers possible negative comments, ratings, and discrimination when presenting products. The speaker says discrimination varies in degree and form, notes that companies described as Muslim have succeeded internationally, and distinguishes between calling a company Muslim and calling it Islamic.
- 3:20:27
Auditing Branches and Worktrees Before Continuing
The speaker asks for an inventory of branches and worktrees containing uncommitted work and checks whether the repository is clean. After some cleanup, the discussion turns to choosing the appropriate workspace for the next milestone and questions why a task was applied to a seemingly unrelated worktree.
- 3:25:36
Preventing a Coding Agent from Mixing Work Scopes
Despite an explicit CLAUDE.md rule to inspect the worktree and branch before starting a new direction, an agent made website-funnel changes within the application work scope. Separately, pulse-feed modifications were attributed to another agent’s mistake, and the agent was asked to remove them from this worktree and prepare the remaining work for pushing and merging.
- 3:28:12
AI as a Mirror of Its Operator
The speaker argues that an AI assistant substantially reflects how its operator directs it, so part of a failure may come from missing understanding or instructions rather than from the tool alone.
- 3:29:26
Trying an Open-Source Voice-Dictation Alternative
The speaker says he uses WhisperFlow and tried a similar open-source tool, finding it functional but still in need of optimization. The other tool's name is uncertain in the captions.
Got a question about something in this stream?
Ask in the group