carlosmarcial

I built the tech stack behind ChatRAG to handle the increasing number of clients I started getting about a year ago who needed Retrieval Augmented Generation (RAG) powered chatbots.

After a lot of trial and error, I settled on this tech stack for ChatRAG:

Frontend

- Next.js 16 (App Router) Latest React framework with server components and streaming

- React 19 + React Compiler: Automatic memoization, no more useMemo/useCallback hell

- Zustand: Lightweight state management (3kb vs Redux bloat)

- Tailwind CSS + Framer Motion: Styling + buttery animations

- Embed a chat widget version of your RAG chatbot on any web page, apart from creating a ChatGPT or Claude looking web UI

AI / LLM Layer

- Vercel AI SDK 5 – Unified streaming interface for all providers

- OpenRouter – Single API for Claude, GPT-4, DeepSeek, Gemini, etc.

- MCP (Model Context Protocol) – Tool use and function calling across models

RAG Pipeline

- Text chunking → documents split for optimal retrieval

- OpenAI embeddings (1536 dim vectors) – Semantic search representation

- pgvector with HNSW indexes – Fast approximate nearest neighbor search directly in Postgres

Database & Auth

- Supabase (PostgreSQL) – Database, auth, realtime, storage in one

- GitHub & Google OAuth via Supabase – Third party sign in providers managed by Supabase

- Row Level Security – Multi-tenant data isolation at the DB level

Multi-Modal Generation

- Use Fal.ai or Replicate.ai API keys for generating image, video and 3D assets inside of your RAG chatbot

Integrations

- WhatsApp via Baileys – Chat with your RAG from WhatsApp

- Stripe / Polar – Payments and subscriptions

Infra

- Fly.io / Koyeb – Edge deployment for WhatsApp workers

- Vercel – Frontend hosting with edge functions

My special sauce: pgvector HNSW indexes (m=64, ef_construction=200) give you sub-100ms semantic search without leaving Postgres. No Pinecone/Weaviate vendor lock-in.

Single-tenant vs Multi-tenant RAG setups: Why not both?

ChatRAG supports both deployment modes depending on your use case:

Single-tenant

- One knowledge base → many users

- Ideal for celebrity/expert AI clones or brand-specific agents

- e.g., "Tony Robbins AI chatbot" or "Deepak Chopra AI"

- All users interact with the same dataset and the same personality layer

Multi-tenant

- Users have workspace/project isolation — each with its own knowledge base, project-based system prompt and settings

- Perfect for SaaS products or platform builders that want to offer AI chatbots to their customers

- Every customer gets private data and their own RAG

This flexibility makes ChatRAG usable not just for AI creators building their own assistant, but also for founders building an AI SaaS that scales across customers, and freelancers/agencies who need to deliver production ready chatbots to clients without starting from zero.

After building multiple RAG powered chatbots that performed well and passed some accuracy tests, I packaged this stack into ChatRAG so developers can ship high quality RAG chatbots faster, whether it's for a personal AI assistant, a multi-tenant SaaS platform, or client projects.

My long term vision is to keep evolving ChatRAG so I can eventually release a fully open source version for everyone to build with.

Thanks for your comment! Totally agree, LinkedIn is a big player with cheaper options. But if you're a high-tech company, sometimes it’s better to go straight to where your ideal candidates are. We’re focused on niche industries like AI, robotics, and quantum computing, so it’s a more direct way to find specialized talent.

Thanks for the great question! I totally agree—job boards live or die by the quality and quantity of jobs.

What makes us different is that we focus specifically on cutting-edge industries like AI, quantum computing, and robotics. If you're Acme Inc., we help you reach top-tier talent looking for those kinds of roles. For someone like Jane Appleseed, it’s about finding meaningful jobs in high-impact fields, not just another generic listing.

We’re trying to create a place where talent and companies in innovative spaces can connect and build the future together. Appreciate the feedback!

Over the past couple of months, I’ve been proving to myself that I can translate my creative coding skills into an indie hacker/maker-type project.

And what does every emerging indie hacker create? That’s right—a job board. But I didn’t want to create just any job board; I wanted to create the job board to end all job boards.

I created Accelerated Jobs to help people find MEANINGFUL work in areas like: AI, Biotech, Fusion, Industrial Automation, Machine Learning, Quantum Computing, Renewable Energy, Robotics, Self-driving Vehicles, Spatial Computing, Space Tech and Thermodynamic Computing

I actually think jobs aren't disappearing, but work is being reimagined—and AcceleratedJobs.com is here to make it meaningful.