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
NexusAI — Enterprise Agentic Assistant
NexusAI is a production-ready enterprise agentic conversational platform that revolutionizes e-commerce through intelligent intent classification and semantic product search. Built with LangGraph, FastAPI, and Next.js 15, this AI assistant utilizes advanced RAG pipelines and Supabase pgvector to deliver highly personalized, context-aware recommendations in real-time. By leveraging Groq's LPU for sub-second LLM inference and FlashRank for cross-encoder analysis, the system ensures unparalleled accuracy and speed. This project highlights my advanced proficiency in LLM orchestration, vector databases, and real-time frontend integration via Server-Sent Events.
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
Spent some time on and off building something just for fun: a live voting site for the argument that never ends — Messi vs Ronaldo.
The football angle is just the hook. Underneath it's closer to a small product than a toy:
- Real payments, real-time — $1 buys a vote, and the global and per-country standings update live as they land
- A 14-tier progression system, Debutant up to GOAT, each level earned and never lost, with three visual card finishes (stands / pitch / history) as you climb
- Every voter gets their own generated card — name, country, camp, vote count — with a shareable page of its own
- A full visual identity built from scratch: stat-heavy and editorial rather than templated
No prizes, no payouts — it's a fan-support board, not a betting site. What I actually wanted to test was whether a dead-simple mechanic (pick a side, back it with money) could carry its own visual language and hold up on repeat visits.
If this one lands, I'd like to keep going with the same format for other sports — there's no shortage of GOAT arguments to settle.
Support and go pick a side: thegoatdebate.io
Might turn this into a proper case study once there's more to show — for now, curious what a design-minded crowd makes of the execution.
Well done! 🚀
One of our clients needed 70 ad video variants in a week, so we built an AI agent to do it in seven hours.
The build itself was a project. It required:
- Assembling a Claude agent connected to Remotion
- Building a knowledge base of the client's brand: brand guide, movement language, hook creation guidelines, past deliverables, briefing structure
- Teaching the agent to read the brief the way our team would
- Composing hook variants adapted for each aspect ratio (1x1, 9x16, 16x9)
- Assembling endcards and outputting files named and organized for QA
The first batch still needed heavy manual editing. The second, the agent assembled alone.
The third came out fully stitched (hook, body, CTA, disclaimer) needing only a review pass before shipping.
Each iteration meant testing what the agent got wrong, adjusting the skills, refining the knowledge base, and rerunning until the output matched the standard our team would ship.
When we sent the batch, the client feedback was: "The revised videos look great. We'll start using them immediately."
To show what this workflow can do, the video in this post was produced by this exact system.
If your business needs to ship ads at this volume, let's talk 👉 apta.agency/contact
70 ad variants in seven hours is wild. remotion?
Heron AI is an agent that reviews architectural models and flags building code problems. The hardest part of the brief was making that legible in the first few seconds.
We built the hero as a small interaction instead of a statement. The visitor drags across a black and white building sketch and violations surface where they drag, annotated with real IBC clauses. It puts the person in the agent's position for a moment.
Everything after that section can then talk in normal product language, because the core idea has already landed without needing to be described.
Really smart way to make a technical AI capability tangible—the drag interaction lets the user discover the value instead of reading a claim. The architectural visual language supports that clarity beautifully.