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
Most chatbots give wrong answers because of retrieval, not the model: what gets pulled in before it ever answers. So I built a RAG chatbot that answers straight from your own documents, with sources you can check. Qdrant vector store, FastAPI + LangChain backend, clean chat UI. Get retrieval right and it stops making things up. Happy to build one of these on your docs. More of my work at kumarsomesh.com.
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
some late work i've been doing with felt👇
quick thought i can't shake: most ai interactions make you translate. you feel something, turn it into words, the ai answers in words, you turn that back into feeling. two translations for something that was never meant to be spoken in the first place.
so with felt i'm trying something different: an interface that just reads what you're feeling and talks back the same way, through touch. no typing, no talking, no middleman.
still early, still prototyping, having way too much fun building this one🤘
Nice work
Generative AI Automotive Diagnostics & Support Platform AI Chatbot
My role. Lead AI Engineer | LLM Integration | Conversational AI Development
Project description.
Engineered a Generative AI-powered automotive support platform that enables users to diagnose vehicle issues, receive maintenance recommendations, and access repair guidance through natural language conversations. Developed AI-driven troubleshooting systems, intelligent recommendation engines, maintenance management workflows, and customer support automation features. The solution provides 24/7 vehicle assistance, helping car owners quickly identify potential issues and make informed maintenance decisions through an intuitive conversational interface.
Skills and deliverables
Automation
AI Chatbot
Conversational AI
Natural Language Processing
Artificial Intelligence
A production multi-tenant AI customer-engagement platform I designed and built end to end. A chatbot answers customer questions 24/7 from each business's own knowledge base (RAG), captures leads, sends instant Telegram alerts, and hands off to a human in real time. Includes a voice assistant and a per-client analytics dashboard.
Self-hosted — each business fully owns its instance, code, and data, with no SaaS lock-in.
Built solo: architecture, backend, AI integration, infrastructure, deployment, and security.
Stack: Next.js, Directus, PostgreSQL (pgvector), Redis, n8n, Twilio, OpenAI / Claude / Gemini, Docker.
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.
MagicPath
The canvas is infinite, and exploration is becoming the workflow. How are you using MagicPath?
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?