You will get a custom RAG chatbot trained on your data — GPT, Claude, cited answers
Rising Talent
Project details
Most AI chatbots make things up. This one does not.
I build RAG chatbots that answer only from YOUR content: your documents, website, or knowledge base. Every answer is grounded in your data and shows the source it used, so your team and customers can trust it.
What you get:
• A chatbot trained on your PDFs, docs, website, or wiki
• Every answer cited to the exact source
• A guardrail: if the answer is not in your data, it says so instead of inventing one
• Powered by Claude or GPT, your choice
• Delivered as a web widget, an API, or connected to your tools
You own everything: source code, database, and infrastructure. It runs under your own API key. No training on your data, retrieval at query time only.
How I work: fast and fully async, with a short screen-recording walkthrough of every delivery, so you see what was built without a meeting.
My stack: Python, FastAPI, PostgreSQL/pgvector, Claude and OpenAI, Redis, Docker. I also build production AI systems, including 3 open-source MCP servers on GitHub. I am earning my first reviews here, so you get my full attention.
Send me a sample of your docs and I will tell you exactly what is possible.
I build RAG chatbots that answer only from YOUR content: your documents, website, or knowledge base. Every answer is grounded in your data and shows the source it used, so your team and customers can trust it.
What you get:
• A chatbot trained on your PDFs, docs, website, or wiki
• Every answer cited to the exact source
• A guardrail: if the answer is not in your data, it says so instead of inventing one
• Powered by Claude or GPT, your choice
• Delivered as a web widget, an API, or connected to your tools
You own everything: source code, database, and infrastructure. It runs under your own API key. No training on your data, retrieval at query time only.
How I work: fast and fully async, with a short screen-recording walkthrough of every delivery, so you see what was built without a meeting.
My stack: Python, FastAPI, PostgreSQL/pgvector, Claude and OpenAI, Redis, Docker. I also build production AI systems, including 3 open-source MCP servers on GitHub. I am earning my first reviews here, so you get my full attention.
Send me a sample of your docs and I will tell you exactly what is possible.
AI Algorithms
Large Language Model, Transformer ModelAI Applications
AI Chatbot, Conversational AI, Natural Language Generation, Natural Language Understanding, Text RecognitionAI Development Language
PythonAI Tools
StreamlitAI Models
ChatGPT, GPT-4What's included
| Service Tiers |
Starter
$129
|
Standard
$349
|
Advanced
$749
|
|---|---|---|---|
| Delivery Time | 4 days | 7 days | 14 days |
Number of Revisions | 1 | 2 | 3 |
AI Model Integration | |||
Batch Normalization | - | - | - |
Database Integration | |||
Detailed Code Comments | - | - | |
Image Upscaling | - | - | - |
MLOps | - | - | - |
Model Deployment | |||
Model Documentation | - | - | |
Model Monitoring | - | - | |
Model Testing & Optimization | - | ||
Model Tuning | - | - | - |
Natural Language Processing | |||
NLP Tokenization | - | - | - |
Pre-Training | - | - | - |
Prompt Engineering | |||
Setup File | - | - | - |
Source Code |
Optional add-ons
You can add these on the next page.
Fast Delivery
+$80 - $200
Additional Revision
+$40
Additional Data Source
(+ 3 Days)
+$120
Admin Dashboard (Streamlit)
(+ 3 Days)
+$150Frequently asked questions
About Armen
AI Chatbot & RAG Developer | MCP Servers, AI Agents | Python & Claude
Bornheim, Germany - 8:50 pm local time
What I've shipped:
- 4 apps published on the Apple App Store (Flutter and Swift/SwiftUI), including AI-powered image and music generation apps.
- A production trading system powered by Claude AI agents, processing live market data with Python, FastAPI, PostgreSQL, Redis, and Docker.
- 3 open-source MCP servers on GitHub (security audits, scholarly papers, live news feeds) that connect Claude to real APIs and live business data. This is a rare, code-verifiable proof almost no freelancer can show.
- Google AI Professional Certificate.
What you get when you hire me:
- AI chatbots and virtual assistants on the Claude and OpenAI APIs, with custom agent orchestration.
- RAG pipelines with document ingestion, vector search, and answers that cite their sources, so your users can trust them.
- Custom MCP servers that give your AI agent live access to your tools, APIs, and data.
- Python backends, REST APIs, and AI-powered iOS, macOS, and cross-platform mobile apps.
How I work: fast, async, and clear. I deliver in tight scoped milestones with a short screen-recording walkthrough of every deliverable, so you always see exactly what was built without a meeting. Your data stays yours: no training on client data, grounding at test time only.
Tools: Python, FastAPI, PostgreSQL, Redis, Docker, Claude API, OpenAI/ChatGPT API, LangChain, RAG, vector search, MCP, n8n automation, Next.js, TypeScript, Flutter, Swift/SwiftUI.
I'm building my first reviews on Upwork, so you'll get my full attention and over-delivery on the first project. If you're working on an AI chatbot, RAG system, MCP server, or automation, send me the brief and I'm happy to answer any questions you may have.
Steps for completing your project
After purchasing the project, send requirements so Armen can start the project.
Delivery time starts when Armen receives requirements from you.
Armen works on your project following the steps below.
Revisions may occur after the delivery date.
Kickoff and data review
I review your documents, plan the chunking strategy, and confirm scope: what sources, what questions users will ask, how answers should look.
Build — ingestion, embeddings, retrieval, UI
I build the full pipeline: document processing, vector storage, retrieval logic, AI answer generation with source citations, and a clean web interface.