I will build a langgraph multi agent ai pipeline for your use case

R
rishibethi
R
rishibethi
Rishi Bethi

About this gig

Are you trying to automate a business workflow with AI but don't know where to start? Or did you start, and now you're stuck with a prototype that doesn't work in production?


I build multi-agent AI pipelines that actually ship.


What I deliver:

- Multi-agent systems using LangGraph (the same framework used by production AI teams at major companies)

- RAG pipelines that answer questions from your documents accurately

- FastAPI backends you can plug into your existing product

- AWS deployment so your system runs 24/7 without you managing servers

- Evaluation scores proving the system works before I hand it over


I have built and deployed a 6-agent compliance pipeline (reguliq.eu) to AWS ECS Fargate with CI/CD, pgvector RAG, and RAGAS evaluation. I also published an open-source LLM diagnostics library with 209 passing tests on PyPI.


Common use cases I have delivered:

- Customer support agents that answer from your knowledge base

- Document processing pipelines (contracts, reports, policies)

- Multi-step research and summarization agents

- Compliance and regulation analysis systems


Message me before ordering so I can confirm your use case fits the package.

Get to know Rishi Bethi

Rishi Bethi

AI Engineer ,LangGraph Multi Agent Systems , RAG Pipelines

  • FromSweden
  • Member sinceMay 2026
  • Languages

    Telugu, English, Hindi, Swedish
I build production-grade AI systems that actually ship. MSc AI & Automation student at University West, Sweden, with 2+ years as a Data Scientist and a live 6-agent LangGraph compliance pipeline deployed on AWS ECS Fargate (reguliq.eu). I specialize in multi-agent orchestration with LangGraph, RAG systems with pgvector and cross-encoder reranking, LLM evaluation using RAGAS, and FastAPI backends. I have published an open-source LLM diagnostics library on PyPI with 209 passing tests. If you need an AI pipeline built correctly, evaluated rigorously, and deployed to production, I can do that.

My Portfolio