Browse categories
Explore
Fiverr Pro
English
$
USD
I will build a custom ai agent with langgraph, rag and fastapi
Eman N
About this gig
Tired of "AI agents" that are just ChatGPT wrapped in a chatbot widget?
I build real agentic systems stateful LangGraph agents that call tools, reason over your data, and get work done. Production architecture, not a demo.
WHAT I BUILD:
Custom AI agents with LangGraph (ReAct, multi-step reasoning, tool-calling)
RAG systems trained on YOUR documents, PDFs, or database
Multimodal agents that read images, receipts, invoices
WhatsApp/chat AI agents with real conversation state
FastAPI backend + Next.js frontend, fully deployed
RECENT WORK:
Multimodal AI agent auditing receipts vision OCR + math-verification to the cent
Stateful WhatsApp AI agent for a skincare brand hybrid RAG search, order tracking, 23 automated tests
Multi-tenant CRM with AI relationship coach + AI-drafted replies
TECH STACK: LangChain, LangGraph, FastAPI, Python, Next.js, TypeScript, Vector Databases, OpenAI
Have a use case or just a messy manual process? Message me. I'll tell you honestly if an agent is the right fix.
Get to know Eman N
Eman N
Full Stack Next js Developer and AI n8n Automation Specialist
- FromPakistan
- Member sinceJul 2024
- Avg. response time1 hour
Languages
Urdu, English, Hindi
Hi, I'm Eman—a Full-Stack Engineer specializing in scalable web apps and smart automations.
I build production-grade SaaS platforms, MVPs, and custom portals using Next.js, TypeScript, Supabase, and Tailwind CSS. Alongside web development, I design powerful automation workflows and AI agents using n8n, webhooks, and REST APIs to connect your apps with tools like WhatsApp, Gmail, and CRMs.
Whether you need a founder-ready web application or seamless workflow automation, I deliver clean, reliable solutions built for long-term growth. Let's discuss your project!
FAQ
Do you build both the backend and frontend?
Yes — FastAPI/Python backend, Next.js/React frontend. Standard and Premium include full-stack delivery.
Can you connect the agent to my own data (RAG)?
Yes — I build Retrieval-Augmented Generation systems that pull answers from your PDFs, database, or documents, with proper vector search — not keyword matching pretending to be RAG.
Which vector database do you use?
Depends on your project — I work with Pinecone, Chroma, pgvector, and Elasticsearch, and recommend the best fit for your scale and budget.
What LLMs do you work with?
Any of them — OpenAI, Anthropic (Claude), Llama, DeepSeek, Mistral, or open-source models via Ollama. The integration process is the same regardless of model, so I can work with whichever you prefer or already have API access to.
Can you deploy it for me?
Yes — Standard and Premium include deployment to your hosting of choice. You get full source code either way; you own everything.
