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
Prompt's marketing team was creating LinkedIn content manually across teams with no shared voice, no tracking, and no central system. I built PostPilot to fix that. Turns a brief into three on-brand post variations in under a minute using Claude AI. Brand voice learning extracts tone from uploaded style guides. Imagen 4 generates visuals matched to each post. Multi-workspace architecture lets teams work independently while leadership oversees everything. 13-tag feedback loop improves output over time. Stack: Next.js 16, Supabase, Claude API, Imagen 4.
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
I built an AI support-ticket router on Replit, then tested it against the real model. Three things broke that no mock would have caught.
A silent failure. The first run sent 33 of 40 tickets to "needs a person". The newer Claude model rejects the temperature setting, and my code swallowed the error, so every call quietly failed. Now every failed call is logged.
The model thinks before it answers. With a tight token limit, some tickets got no answer at all and some draft replies stopped mid-sentence. Both have room now, and a cut-off draft is never saved.
One incident looked like four bugs. Four Safari checkout failures each got a different issue key, so the "3+ tickets, same problem" alert never fired. Sorting tickets in arrival order and asking for keys that name the problem rather than the customer turned them into one flagged incident.
If you're adding an LLM feature to your product, test it on real inputs early. The failures are rarely where you expect.
Live demo: https://sorta--krishnanbhupath.replit.app
Scroll → a camera flies through a watch gallery → zooms into the dial → you pick a dial colour → add to cart.
That's the homepage of AUREL, a luxury watch store I designed and built as a portfolio project. ⌚
Under the hood:
• 3D scroll story with React Three Fiber, GSAP and custom shaders
• Full shop flow: filters, product variants, cart, 4-step checkout
• Mobile CSS 3D fallback + reduced-motion layout
• Accessible, fast, and ready to plug into Shopify
Stack: React, Vite, Tailwind, Three.js, GSAP, Zustand
I'm a web developer (React · WordPress · PHP, 6+ years) looking for remote frontend or WordPress roles. Recruiters and hiring managers, my DMs are open.
#ReactJS #FrontendDeveloper #ThreeJS #OpenToWork #WebDevelopment
This is nice
🚀 Created realistic AI-generated UGC videos for PactFresh, blending product-focused storytelling with authentic visuals to create engaging marketing content.
#PactFresh #AIUGC #AIVideoGeneration #ContentCreation
The specific mention of Higgsfield is interesting because it is not the standard tool most people assume for this workflow. Curious how you balanced the product accuracy against the natural movement in those frames to keep the UGC feel authentic.