The network for creativity
Join 1.25M professional creatives like you
Connect with clients, get discovered, and run your business 100% commission-free
Creatives on Contra have earned over $150M and we are just getting started
Back to feedPost
From hypothesis to decision, in one system of record.
We built ExperimentOps, a full-stack A/B testing and decision platform for product and growth teams. Setup, variant assignment, monitoring, statistical analysis and the final ship/kill call live in one workspace, with an audit trail of why every change shipped.
Under the hood:
→ Deterministic SHA-256 bucketing: a user always sees the same variant
→ PostgreSQL row-level security: each tenant is isolated in the database itself, not just the app
→ Analysis snapshots locked to every decision
Built with Next.js, FastAPI and PostgreSQL. Spec to production in 10 phases.
Building a SaaS product or internal tool? Let's talk.
The network for creativity
Join 1.25M professional creatives like you
Connect with clients, get discovered, and run your business 100% commission-free
Creatives on Contra have earned over $150M and we are just getting started
Related posts
🎃🎃🎃 Halloween special, part 1: the scariest things that happen to apps after launch
🕯️ The App Store rejection email. It lands the morning of your big announcement
🔔 The permission prompt asked too early. One tap on "Don't Allow" and push notifications are gone for good
🕸️ The slow screen. Nobody complains. Users just quietly stop opening the app
🧟 The forgotten test account. It breaks, and App Review rejects you for something you never saw
🪦 The abandoned app. No updates for a year, and the OS update that finally kills it
The antidote is boring and works every time: test early, monitor everything, ship small.
Which one scares you most?:)))
Love this!
I built an AI-powered workflow that helps real estate professionals turn property details into polished listings for websites, property portals, and social media. Using Gemini and n8n, it generates the content, highlights key selling points, and flags missing information. Before anything is saved to Airtable, the listing is sent via WhatsApp for approval.
Tools: n8n, Google Gemini, WhatsApp Business, Airtable
Having the listing go to WhatsApp for approval before anything lands in Airtable keeps a human in the loop without slowing the flow down. When Gemini flags missing property info, does the approver fill the gaps in WhatsApp, or does it loop back through n8n for a regenerate?
