Senior Machine Learning Engineer

Munich Germany Ref: 2021

About Our Client

Our client is building a no-code, agent-oriented AI platform designed for organisations that rely heavily on technical documentation, operational records, and complex knowledge workflows. Their platform specialises in automating high-value decision processes that have historically been locked inside disparate document stores, legacy systems, and deep tacit expertise.

Their multi-agent architecture enables sophisticated reasoning over large, evolving bodies of unstructured information, turning what used to be manual, consulting-driven processes into scalable, intelligent, software-powered operations.

They collaborate with major enterprises across Europe, North America, LATAM, and Asia.

The Role

In this role, you will lead the development of the document?centric reasoning engine that powers our client’s agentic AI platform. Your work sits at the intersection of NLP, retrieval-augmented generation (RAG), knowledge-graph construction, and multi-agent inference. You’ll design and implement pipelines that convert raw document collections into structured, queryable, and semantically rich knowledge assets.

You’ll focus on robust grounding, factual consistency, and domain?aware retrieval rather than generic LLM generation. Your systems will enable autonomous agents to interpret documents, resolve context, connect concepts across sources, and provide trustworthy insights that reflect real-world constraints and operational logic.

Responsibilities

Document Understanding & Semantic Parsing

• Build advanced NLP pipelines for extracting entities, relationships, constraints, and procedural details from large, heterogeneous document sets (PDFs, CAD specs, procedural manuals, contracts, reports).

Retrieval?Augmented Generation (RAG) Systems

• Architect scalable RAG workflows using hybrid search (semantic + sparse), document chunking strategies, embedding optimisation, and grounded generation approaches tailored to enterprise knowledge.

Knowledge Graph Construction & Enrichment

• Automate the transformation of document-level insights into structured knowledge graphs, capturing dependencies, hierarchies, workflows, and cross-document relationships.

Agentic Reasoning Over Documents

• Develop agent-based inference pipelines that allow AI agents to perform multi-step reasoning, validation, and decision-making using retrieved evidence rather than synthetic generation.

Contextual Affordance Mapping

• Implement systems that identify actionable opportunities, risks, requirements, or constraints embedded within textual content, enabling downstream agents or users to operate with domain?aware intelligence.

Optimisation & Production Deployment

• Bring state-of-the-art NLP and retrieval research into reliable production workflows that deliver measurable efficiency and quality gains for customers.

Requirements

• MS/PhD in Computer Science, NLP, Information Retrieval, or a related field
• 3+ years of industry experience in large-scale NLP or IR systems
• 2+ years leading projects in RAG, document intelligence, or knowledge-graph-based reasoning
• Experience with vector databases, hybrid retrieval, embedding optimisation, and LLM orchestration frameworks
• Familiarity with agentic systems (GraphRAG, LangGraph, LlamaIndex, etc.) is a strong advantage
• Expertise in using libraries such as spaCy, HuggingFace Transformers, Haystack, LangChain, or similar
• Strong understanding of scalable, production-grade software systems
• Prior experience in a high-growth, fast-moving tech environment is beneficial
• Excellent English communication skills

Why Work With Our Client

• Push the boundaries of enterprise-grade RAG systems and document-grounded reasoning
• High ownership and autonomy , influencing how intelligent agents process and understand complex knowledge
• Collaborate with an international team of NLP, knowledge-engineering, and software experts
• Strong growth potential with competitive compensation and long?term value creation