Browse categories
Explore
Fiverr Pro
English
$
USD
I will build a custom claude ai agent or multi agent system
Dennis Janssen
About this gig
Not a chatbot. An AI that thinks, plans, and delegates.
I build custom AI agents and multi-agent systems using the Claude API (Anthropic). These are autonomous systems that break down complex tasks, use tools, call APIs, and route work to specialized sub-agents all without human input.
This is not a wrapped ChatGPT. This is a purpose-built agent architecture designed around your specific workflow.
What I build:
- Single agents with tool use (web search, database, API calls, file reading)
- Orchestrator + sub-agent teams (one agent directs, others execute)
- Agents with memory and persistent context
- Agents connected to MCP servers (Notion, Slack, GitHub, etc.)
- Internal agent systems (research, writing, classification, routing)
Real examples:
- Marketing agent team: strategist writer brand checker publisher
- Operations agent: reads inbox classifies routes drafts replies
- Research agent: searches web summarises writes report stores in Notion
- Sales agent: qualifies lead scores assigns drafts outreach email
Stack: Claude API · Python / TypeScript · MCP · Tool use · Vercel / Railway
Get to know Dennis Janssen
Dennis Janssen
Build AI Tools, Websites and Workflows
- FromNetherlands
- Member sinceJun 2026
- Avg. response time1 hour
Languages
Dutch, English
I build websites, AI tools, automated workflows, and custom AI agents for businesses that want to move faster. Based in the Netherlands, working globally.
I've built automation pipelines with n8n, custom AI assistants with the Claude API, modern websites with Next.js and Astro, and business-ready AI products from scratch.
If you have a manual process that eats time, a product idea you want to validate fast, or a website that should convert better — let's talk.
Skills: n8n · Claude API · Next.js · Astro · Python · REST APIs · AI Agents
My Portfolio
FAQ
Q: What is a multi-agent system?
A: Multiple AI agents working together — one orchestrator breaks down the task and assigns it to specialized sub-agents. Each agent has a specific role. The result is more accurate and scalable than a single agent trying to do everything.
Q: How is this different from a regular chatbot?
A: A chatbot responds to questions. An agent system takes actions , it searches, writes, routes, stores, and reports autonomously. You define the goal; the agents figure out the steps.
Q: Which MCP servers can you connect?
A: Notion, Slack, GitHub, Gmail, Google Drive, Linear, Airtable, and any custom API. If it has an MCP server or REST API, it can be connected.
