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
Designed and developed an enterprise AI knowledge platform using Java, Spring Boot, Spring AI, React, and PostgreSQL with PGVector.
The platform enables organizations to upload documents, perform semantic searches, and ask natural-language questions against their own knowledge bases using Retrieval-Augmented Generation (RAG).
Key capabilities
Document ingestion and vector embedding generation
Context-aware AI answers grounded in retrieved documents
Source citations for transparency and verification
Secure authentication using Keycloak
Multi-tenant data isolation
Integration with local and cloud-hosted LLMs
REST APIs, observability, and automated testing
Technology stack: Java 17, Spring Boot, Spring AI, React, TypeScript, PostgreSQL, PGVector, Keycloak, Ollama, OpenAI, and AWS.
My role: Solution Architecture, AI Engineering, Backend Development, and Technical Leadership.
This project demonstrates how businesses can transform fragmented documents into searchable, AI-powered organizational knowledge while maintaining security, scalability, and control over their data.
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
TypeScript 5 → 7 upgrade for an open-source TypeScript CLI tool, done in two steps (5 → 6 → 7): compiler option changes, new type errors fixed, tooling updated. 1177/1177 tests passing before and after, build and type check green, CI parity table in the PR. The maintainer closed it to fold the change into their own upcoming PR — the diff and verification are public in the link.
5 to 7 in one go? brave. which compiler option bit you the most
Is this really happening?
After 17 years as a graphic designer and illustrator, I find myself, at 43 years old, feeling forced to pursue a degree in Artificial Intelligence Engineering.
And honestly, I'm torn.
On one hand, the thought of going back to university, with everything it demands, feels overwhelming. I'm afraid of losing my livelihood, despite having spent years proving myself, consistently pushing beyond expectations, standing out, and becoming exceptionally good at what I do.
Yet somehow, fear is eating away at me.
The realization that everything I've built over nearly two decades could be hanging by a thread is terrifying.
But on the other hand, I'm genuinely excited.
The idea of understanding how to build AI, how it thinks, how it learns, and potentially making a living from that knowledge fascinates me.
Applied mathematics. Programming. Computer science fundamentals. Algorithms. Machine learning.
An endless journey of studying, practicing, failing, learning, and starting again.
At 43.
As a man, I feel a responsibility to protect and provide for my family.
But as a human being, I feel an even deeper responsibility to contribute, however small my contribution might be, to preventing the darker consequences of the technology we're creating.
Mass unemployment. Surveillance and control. Suicide. Hunger. The loss of human identity. New threats. Wars. Conflicts we can barely imagine today.
We are building something that could fundamentally reshape what it means to be human.
And while the possibilities are extraordinary, so are the risks.
I don't want to spend the rest of my life watching from the sidelines, afraid of what comes next.
I want to understand it.
I want to be part of it.
And above all, I want to help make sure this technology serves humanity rather than destroys what makes us human.
The future isn't written yet.
It's in our hands.
Let's do something about it.
Nexus — AI Infrastructure & Compute Platform Landing Page
An enterprise-grade web dashboard ecosystem engineered for machine learning developers and AI engineering teams. The workspace provides complete observability over GPU compute allocations, training telemetry, serverless inference routing, and real-time model benchmarks.
Project Goal
The objective was to eliminate the friction of chaotic terminal logs and disconnected telemetry tools by building a unified, closed-loop AI infrastructure system. Every interface card was crafted to maximize visual scannability, streamline cluster-scale workflows, and deliver instant operational clarity without cognitive overload.
The full preview landing page is available to check directly inside the case study link! 👉🏻 Check it out
