PandaCodeGen reference
PandaCodeGen Project Evidence
Every published project, its relationship to PandaCodeGen, and what is and is not claimed about each. Founder-owned properties are labelled as such and are not presented as independent client proof.
Every value on this page was checked against source on 8 August 2026. Figures that can change carry the date they were checked. This page states facts about PandaCodeGen only.
Across 1 March to 18 August 2026 this store recorded 760 orders worth $271,620.61, of which 63 orders and $16,783.76 were attributed to AI assistants, 6.2% of revenue. ChatGPT produced $11,614.78 against Google Ads $1,733.35. Attribution combines referrer capture with a self-reported source question at the point of enquiry, because referrer alone undercounts AI referrals; Google Search Console’s Generative AI report for the same property independently recorded 48,900 impressions over three months. Lead-to-order conversion rates are deliberately not published because the lead and order records do not share a window.
Every page, product and blog post migrated to a custom Next.js build in about 22 days, with Search Console crawl validation before DNS cutover. Performance and hosting-cost figures for this project are withdrawn pending reconciliation of the original test records.
Migrated to a custom Next.js storefront. Owned by co-founder Imran Raza Ladhani, so it is labelled as a founder-affiliated project rather than presented as client evidence.
An operations platform with role-based access, payments, attendance and twenty Supabase Edge Functions, built in 10 weeks.
Custom build. Published without performance or traffic claims, because the site is too new for that evidence to exist.
Migration to a self-manageable CMS. No before-and-after performance figures are published, because the previous platform baseline was never captured.
How evidence is published
| Press | Co-founder and Lead Engineer Hassan Jamal was quoted in Woman's World on 29 May 2026, in an article about remote AI-training work. It is the company's only press placement to date. |
|---|---|
| Evidence policy | Project results are published with the relationship, date, measurement method and stated limitations. Founder-affiliated projects are labelled as such and are not presented as independent client proof. |
| Review sourcing | Client reviews are published as named individual excerpts linked to their source page, never as a star rating or an aggregate score. As checked on 17 August 2026, four of the reviews at trustpilot.com/review/pandacodegen.com carry Trustpilot’s own “Unprompted review” label, which Trustpilot applies when the business did not invite the review. The Clutch review is identity-verified by Clutch and is not labelled unprompted, because Clutch does not publish that label. |
| Regulated data | PandaCodeGen will execute a Business Associate Agreement where a project involves protected health information under HIPAA. Signing a BAA defines responsibilities for that engagement; it is not a certification and is not a claim of blanket HIPAA compliance. |
| Crawl policy | All crawlers, including AI crawlers, may access the site. Only /api/ and /_next/data/ are excluded. The machine-readable version is at /robots.txt. |
| Corrections | Mutable facts carry the date they were checked. Where a published figure is later found wrong it is corrected rather than quietly removed, and the editorial policy states how. |
Two entries above carry no performance figures at all. That is deliberate: one site is too new for that evidence to exist, and on the other the previous platform baseline was never captured before switchover. Reconstructing a before-and-after after the fact would be a guess presented as a measurement.
This is one of the PandaCodeGen reference pages. The full set is indexed on the company reference page.