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
Taskadar — AI-powered local-services marketplace
The problem: local-services marketplaces usually force users through rigid category trees and filters before they can even describe what they actually need.
What I built: a natural-language search where you type the task in plain language and the AI extracts trade, urgency and location, then ranks nearby providers by rating, availability and price band with a friendly summary. Full buyer flow from search to ranked results to provider profile to booking confirmation, plus a two-sided marketplace where sellers publish listings that become instantly searchable. Bring-your-own-key for Gemini 2.5 Flash, Claude Opus, or GPT-4o mini, with a built-in keyword matcher as a graceful fallback when no key is set. One codebase runs on iOS, Android and web.
Stack: Expo SDK 54, React Native 0.81, TypeScript (strict), Expo Router, React Native Reanimated.
Live demo: taskadar.vercel.app
Source code: github.com/nabilsaiyan/taskadar
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
Trending
retrofuturism
Share your favorite designs from the past exploring the distant future. Or your own take on the aesthetic 🚀
halftone
Dots. Texture. Grain. Let's see your designs using the halftone technique.
Midjourney
Creatives are showing off what they're making with Midjourney.
colorgrade
Video editors sound off: How do you enhance the mood of your video in color grading?
Related posts
DayMate is a lightweight AI-powered daily planning assistant built with React, Node.js, Express and Groq AI. Users can add, complete and delete their daily tasks. When they click 'Plan My Day with AI', the React frontend sends the tasks to an Express backend through a POST API. The backend securely handles the Groq API key using environment variables and sends the user's tasks to an LLM with a structured prompt. The AI generates a short and realistic plan for the day, which the backend sends back to React and displays in the UI.
Love this AI daily planner concept! Using React, Node, and Groq together makes it super powerful
Excited to officially start offering my services on Contra!
I’m a Full-Stack & AI Engineer with 6+ years of experience, helping startups and businesses turn ideas into scalable, production-ready products.
I specialize in building:
🤖 Custom AI Web Applications
⚡ AI Agents & Automation
🚀 AI SaaS & SaaS Products
💬 AI Chatbots & RAG Applications
🌐 Full-Stack Web Applications
🔗 APIs & Backend Systems
📊 Custom Business Dashboards
My core stack includes Python, FastAPI, Django, React, Next.js, Node.js, MERN, LLMs, Machine Learning, and modern AI technologies.
I also run a YouTube channel with 38K+ subscribers, where I teach AI, Python, Machine Learning, and software development.
I’ve just launched my first Contra service: Custom AI Web Apps & SaaS Development.
Have an AI or SaaS idea you want to turn into a real product? Let’s build it. 🚀
#AI #FullStackDevelopment #SaaS #AIDevelopment #Python #FastAPI #NextJS #ArtificialIntelligence
HOO POS — Custom Point-of-Sale Platform
Full disclosure: I don't come from a programming background. I actually got accepted into Architecture and IT programs but ended up picking Visual Communication Design because my friends were there. Fast forward, I run a family retail store and instead of hiring a dev team or buying an off-the-shelf POS, I decided to just... build one myself.
Not by learning to code line-by-line the traditional way, but by directing AI tools like an architect directs a construction crew. I know exactly what the system needs to do (because I run the store that uses it), I make every architectural and product decision, and I push the build forward daily. The "how" of getting AI to actually execute that vision reliably that's kind of my own thing I've figured out along the way.
One thing that's been key: I treat documentation as part of the build itself, not an afterthought. Every sprint, every architectural decision, every "why we did it this way instead of that way" gets written down as I go. It's less about writing a manual for other people and more about keeping the whole system honest so a month from now I (or the AI) can pick up exactly where things left off without re-explaining the entire history. Turns out that habit is doing a lot of heavy lifting on a project this complex.
What it actually does:
Multi-unit inventory conversion (stock arrives per box, sells per piece math handled automatically)
Tiered pricing & supplier debt tracking with full audit trail
Offline-first — keeps running even when the internet cuts out, syncs later
Blind cash close so shift handovers stay honest
Bilingual (ID/EN), role-based access, automated cloud backup
A dashboard that doesn't just show charts it tells you in plain language what actually needs your attention today
Stack: React + Zustand frontend, Node/TypeScript backend (modular route architecture), Prisma + PostgreSQL, deployed on Railway with versioned API, IndexedDB for offline reliability.
Status: Live in production. Real transactions, real cashiers, real pressure every day not a demo.
Why this matters: I'm not building for a portfolio piece that just needs to demo well once. I'm building for a business I run, where a bug means someone can't check out a customer. That constraint changed how I approach everything no cutting corners on things like data integrity or offline handling, because I'm the one who'd get the angry call.
