Job Description

Rud Pedersen is looking for a hands-on AI Automation & Platform Engineer to build and improve workflows, integrations, automations, and platform components and the wider AI transformation programme.

The role focuses on n8n workflow automation, API integrations, document and transcript processing, AI-assisted research, notifications, approvals, Microsoft ecosystem integration, and practical AI tools for public affairs and strategic communications teams. The person should be comfortable building real systems, testing them with realistic data, documenting them properly, and supporting them after launch.

This is not a prompt-only role and not a pure ML research role. Slide decks are not a deliverable unless the workflow actually works afterwards.

This role is based in our Bulgaria office and you must have the existing right to work in Bulgaria. You will be working with colleagues across all our European offices.

What you will do

You will design and build n8n workflows that connect internal tools, documents, AI models, databases, forms, email, Microsoft 365 systems, notifications, and other APIs. You will turn recurring manual tasks into maintainable flows.

You will work on AI-enabled workflows such as meeting transcript summarisation, workshop output capture, stakeholder research, policy monitoring, document classification, proposal support, knowledge retrieval, client-delivery support, approvals, and internal operational processes.

You will help extend and integrate our platforms where needed. This may include file-handling workflows, model/provider integrations, backend services, prompt or assistant patterns, usage analytics, RAG/knowledge retrieval, and connections between AI and other internal systems.

You will work closely with the Technical Product Owner to turn prioritised requirements into working releases. You will test workflows, handle errors properly, document what has been built, and make sure the result can be maintained without heroics.

What good looks like

After three months, you have shipped several useful workflows or platform improvements that teams actually use, with clear documentation, sensible error handling, and visible time saved. Our internal platform has stronger integrations and fewer “clever demo, now what?” moments.

After six months, Rud Pedersen has a reusable library of automation patterns for research, documents, workshops, proposals, and internal operations, plus a more reliable internal platform that can support repeatable work across teams.

Essential experience

• Practical experience building automations in n8n. Strong Make, Zapier, or Power Automate experience is useful, but n8n experience is preferred.
• Good understanding of APIs, webhooks, JSON, HTTP requests, authentication, permissions, secrets management, and error handling.
• Experience using LLMs in workflows, including summarisation, classification, extraction, routing, prompt design, and quality checking.
• Ability to work with documents, forms, spreadsheets, email, transcripts, and structured/unstructured business data.
• Good technical skills in at least one of Python, JavaScript/TypeScript, Node.js, .NET, or similar.
• Ability to work with Linux servers, containers, logs, and production troubleshooting.
• Awareness of secure handling of sensitive business data.
• Strong English communication skills.

Useful additional experience

• Azure, Azure DevOps, GitHub Actions, Docker, logging, monitoring, and production operations.
• Microsoft Graph, SharePoint, Teams, Outlook, OneDrive, Entra ID, or Microsoft Fabric.
• LibreChat, Open WebUI, LangChain, LlamaIndex, Semantic Kernel, or similar LLM application frameworks.
• RAG, embeddings, vector search, document processing, and evaluation workflows.
• Databases such as PostgreSQL, MongoDB, Qdrant, pgvector, Weaviate, or Pinecone.
• Experience with governance, audit trails, approvals, and safe handling of sensitive data.
• Experience in consulting, professional services, public affairs, communications, legal, policy, or research-heavy environments.