(Senior) AI Engineer (m/f/d) in München

Sinpex - Built for people who think one step ahead.

At Sinpex, we don’t just keep up with compliance - we redefine it. We believe KYB (Know Your Business) isn’t a regulatory hurdle; it’s the foundation for businesses to scale with confidence. By transforming complex compliance processes into seamless, intelligent workflows, we empower companies to move faster, stay secure, and build lasting trust.

We’re on a bold mission to become the undisputed global leader in KYB SaaS, simplifying compliance and redefining how businesses establish trust. If you’re ready to push boundaries, make a real impact, and grow with an ambitious team, let’s build the future of compliance together.

We are looking for a (Senior) AI Engineer (m/f/d) to join our team. You build the path from unstructured sources to structured company data: LLM pipelines that pull fields from documents and register extracts, web search for public sources, and a knowledge base filled and curated with AI. You build with existing models rather than doing research of your own.

What you will do:

1. Building LLM pipelines that pull structured fields from documents and register extracts, with quality measured per field and country through eval sets
2. Finding, ranking and verifying public sources on a company, and connecting the search provider
3. Defining the data model of the knowledge base and building AI agents for research and validation
4. Keeping existing register connectors running and modernising them step by step
5. Being the engineering voice in discovery with the product owner and design, and turning technical unknowns into epics and stories
6. Reacting fast to urgent customer issues

What you will need:

1. You measure before you claim: you build a labelled test set for a new document, measure precision and recall per field and notice when a new prompt makes something worse
2. You bring things live and keep them stable: solid Python, your own tests and deployments, logging and monitoring for your pipelines
3. You have run systems in production yourself and fixed failures in operation
4. You have put structured output with LLMs on real PDFs into production and know OCR, layout models and vision LLMs apart
5. You have run RAG in production, know chunking, embeddings and hybrid search, and have measured retrieval quality
6. You know that in KYB a wrongly extracted beneficial owner costs more than an empty field, and you build in confidence values and human review
7. You break unclear problems into stories yourself, document decisions and review thoroughly
8. You handle sensitive data such as IDs and customer documents with care
9. You enjoy working with Customer Success and Compliance
10. You speak and write English, the working language of this team
11. You do not need to know the domain or our knowledge base on day one; you will learn it with us in three to six months. Experience in ML research or model training is not needed.

What you will get:

1. Ownership and Trust: We rely on your commitment and responsibility from day one.
2. Collaborative Environment: You'll work in flat hierarchies and have the opportunity to collaborate closely with the founders and the management team.
3. Dynamic Work Environment: You'll see the impact of your work immediately, allowing you to learn and grow quickly.
4. Flexible Working Hours: We value results over rigid working hours, allowing you to work during your most productive times.
5. Hybrid Working: We value personal meetings but also recognize the value of remote work.

Information about our Process:

This is Important to Us:

We are an employer who respects the principles of equal opportunities and diversity. Individuals of different backgrounds are welcome to join us in shaping the future together. Our company does not discriminate against anyone based on characteristics such as race, religion, skin color, national origin, gender, sexual orientation, age, marital status, or disability.

#J-18808-Ljbffr