Build a Full-Stack Platform with Supabase PostgreSQL Expertise

Build a Full-Stack Platform with Supabase PostgreSQL ExpertiseBuild a Full-Stack Platform with Supabase PostgreSQL Expertise

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

Berckan Guerrero

pro

• Feb 8

Where Are They? (dondeestan.cl): Full-stack platform built on Supabase PostgreSQL from the ground up. Designed and implemented the complete database architecture: 63 migrations, Row Level Security (RLS) policies, full-text search with Spanish stemming, and incremental data sync using SHA-256 change detection. The project aggregates 3,216+ records from 4 external sources through a Python ETL pipeline with 99% deduplication accuracy, AI-powered image restoration (Real-ESRGAN + GFPGAN), and a Nuxt 3 frontend. Hosted on GitHub with a monorepo structure, CI/CD via GitHub Actions, and Vercel deployment. This project demonstrates deep Supabase expertise: auth, storage, database design, RLS, edge functions, and seamless integration with a modern frontend stack.

Completed work

Full Stack Development Supabase Database Development Data Engineering Nuxt.js Python

Ajilo Aanuoluwapo

• Mar 9

From Chaos to Clarity: GhostRank AI UI/UX (In Progress)

Building the future of founder-led content, one pixel at a time.

Founders don’t have time for clunky tools or generic AI. That’s why I’m crafting GhostRank AI with a relentless focus on clarity, speed, and delight.

The onboarding is frictionless, dark mode for late-night inspiration, and a Kanban board that turns content chaos into a visual flow. The “Voice Vault” radar chart is the showstopper: founders see their unique tone mapped out, making every AI-generated post feel unmistakably theirs.

Currently in Progress: I’m refining microinteractions, perfecting the drag-and-drop experience, and tuning the viral score predictor for maximum impact. Every design decision is about empowering founders to own their narrative, without the agency price tag.

AI Automation AI Development Lovable N8N Supabase Full Stack Development

Mani Sharan Bommakanti

• Mar 8

Built an end-to-end NLP pipeline to classify text reviews as Positive, Negative, or Neutral. Used NLTK for tokenization and lemmatization, TF-IDF for vectorization, and trained 3 ML classifiers — Logistic Regression, Naive Bayes, and SVM. Visualized sentiment trends and model accuracy using Matplotlib and Seaborn.

Data Analysis Data Engineering Python scikit-learn Machine Learning Engineering NLTK

Ben Soumahoro

• Mar 11

Great! can you embed you model in a web app?

Robin Beechey

pro

• Mar 9

My first time contributing design work to an open source project. Huge thanks to the Django Debug Toolbar team for trusting me to experiment and explore a visual identity for the project. It was a fun process working with a small team that has been maintaining this tool for years. The identity is built around visibility, illumination, and precision. I explored simple forms like rays, borders, and precise layouts to reflect the idea of revealing what is happening under the hood. Django Debug Toolbar is a long running tool that surfaces request data, SQL queries, and performance information directly in the browser while developing a Django app.