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
Introducing live RAG ,VIXIT Cortex: Enterprise AI That Turns Your Data Into Instant Answers
Category: AI & Machine Learning / Web Development
Tools Used: Node.js, PostgreSQL, Vector Embeddings, REST API, Natural Language Processing, Zero-Trust Security and many more stuff
Project Description:
The Problem: Data-heavy teams — hospitals, law firms, banks, insurers — spend weeks extracting insights from their data. SQL queries, data teams, manual reports. Slow, expensive, error-prone.
The Solution: VIXIT Cortex is an enterprise AI agent that lets teams ask questions in plain English and get instant answers. No SQL. No data teams. No waiting.
What I Built:
Natural language query engine that understands domain-specific terminology
Vector embedding system that processes 18 years of data in under 3 minutes
Real-time analytics with sub-second query response (1.2s latency)
Zero-trust security architecture (HIPAA, SOC 2 Type II, GDPR compliant)
Tenant-isolated PostgreSQL databases with SHA256 token hashing
Human-in-the-loop oversight for critical decisions
Export capabilities (PDF, Excel, CSV) with source citations
Key Features:
Ask questions like "How many patients have eye power above 8?" and get instant answers
Multi-step reasoning that breaks complex queries into logical steps
Persistent memory that remembers context across conversations
99.99% uptime SLA with 24/7 priority support
Integration with existing systems (Postgres, MySQL, Snowflake, BigQuery, Epic, Cerner)
Banger
Thanks 🙌
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 got tired of two things: the nightly "what should I watch?" paralysis across a dozen streaming apps, and rating systems where a single score doesn't really tell you enough about a film.
So I built Quartile (quartilefilm.io), a film discovery and rating platform. The core idea: instead of one ambiguous score, every film is broken into the five elements that actually matter — Plot, Acting (Narration for docs), Cinematography, Novelty, and Ending — each scored on a 1–4 scale: Well Below Average, Below Average, Above Average, Well Above Average. There's deliberately no neutral "average" middle — it forces you to actually take a position on each category instead of lazily defaulting everything to the middle, which keeps the ratings honest.
Those five combine into a consistent score out of 10, but you can also see what each category scored, which helps you understand why a film is rated what it is. Sometimes a movie is really good but the ending or the acting is terrible, so the overall score isn't that high — and without the Quartile system you'd have no way to know that, because a single number can't tell you. That's exactly why I built it: I wanted to know why a film was rated what it was. Essentially it's a formula and some real quality control for how a film gets rated.
The coolest part is it lets you filter in really cool ways. You can filter by Top in any category — so Top Cinematography shows only films rated Well Above Average (4) there. Top Novelty is one of my favorites, and one of the most unique filters we offer. You can also stack these with genre — so Top Novelty in horror, or Top Cinematography in sci-fi — which is where it gets genuinely useful for finding something specific to your mood.
On top of the rating system you can log films you've seen, build and share playlists, and follow other users to see what they're watching. The same way you follow friends with good taste in music to find new songs on Spotify, film is a matter of taste too — you can follow friends with good taste in certain genres to find movies you'll actually like.
It's in beta right now. Would love feedback from this crowd and general thoughts. Of course I'd be stoked if you made a profile and started rating movies and building playlists too — I'm going to grandfather in the first active users and give them something really cool on their profile eventually to mark that they were one of the first Quartile users.
How do the five categories mathematically combine into the final score out of 10. Is it a straight average, or are certain elements weighted higher?
Meet AI is a video conferencing SaaS where AI agents join live meetings as active participants and respond in real time. I built the full application stack.
The hard part is the real-time layer. The backend provisions an AI agent the moment a meeting starts and keeps it running for the full call. Speech runs through a transcript pipeline and generates responses through the OpenAI API in near real time, so the agent replies while the conversation is still moving, not minutes later. After the meeting, the backend writes a structured summary through the same API, handled async by Inngest so nothing blocks the live app.
The stack is Next.js 15 and React 19 with the Stream Video and Stream Chat SDKs, tRPC and Drizzle ORM on Neon PostgreSQL, and Better Auth across the stack.
The platform runs 30 simultaneous live sessions, each with its own active AI agent, with no backend slowdown.
What the client gets: live AI agents that take part in real meetings, plus automatic summaries, on a stack built to hold many sessions at once.
Perfect
Trending
Claude
Claude has entered the design space. How are you using Claude Design?
Contra University
Learn from expert creatives how to earn more using next-gen AI tools.
fifaworldcup2026
The World Cup is here and the whole world's watching. How are you designing for the world stage?
creativeaiflow
Creative AI workflows are evolving. What tools do you use, and what are their strengths and weaknesses?
freelancerlife
Freelancer life is wins, pivots, and everything in between. What’s yours right now?
