Who we are
At Graphwise, we're helping organizations unlock the full value of their data in the age of AI.Formed through the merger of Ontotext and Semantic Web Company, Graphwise combines decades of expertise in knowledge graphs, semantic technologies, and AI. Our platform helps enterprises transform fragmented data into connected, trusted knowledge that powers better decisions, smarter search, and more reliable AI solutions.With a global team of 200+ professionals across Europe, North America, and APAC, we work on complex, meaningful challenges at the intersection of data, AI, and innovation.If you're excited by cutting-edge technology, continuous learning, and solving real-world problems, you'll feel right at home at Graphwise.
About the Job
At Graphwise, we combine state-of-the-art semantic web technologies with industry-specific data and deep process expertise to build the next generation of Graphwise’s GraphRAG and Graph Automation products. Agentic AI based on Large Language Models (LLMs) is transforming how users interact with their business applications, but pre-trained LLMs have a limited understanding of structured and unstructured business data – such as data models, process metadata, and documentation – which makes using agents at scale for complex tasks challenging. Our mission is to overcome this challenge by using Knowledge Graphs (KGs) to address LLM issues such as hallucination, explainability, and compliance.
We are looking for an Engineer who can bridge the gap between scientific research and production AI engineering. You will work at the intersection of knowledge graphs, LLMs, and workflow automation to build scalable, explainable pipelines that power graph retrieval-augmented generation, semantic enrichment, multi-hop reasoning, and automated graph construction (ontologies, taxonomies, and knowledge graphs). Please note that this is not a data science or applied AI position; you need to be an engineer who can ship features with a model.
What You Will Do
- Contribute to the technical direction and productization of a composable, multimodal GraphRAG and Graph Automation architecture that supports unstructured, semi-structured, and structured business data (including code and text).
- Translate advances in LLMs, vLLMs, multimodal learning, and knowledge graphs into a clear technical roadmap, aligning R&D researchers and software engineers on hypotheses, prototyping, acceptance criteria, and architectural decisions.
- Design, optimize, and implement scalable data ingestion, transformation, and enrichment pipelines, as well as caching and state management.
- Integrate various language models – such as encoder-only models, decoder-only generative LLMs, and embedding models – into production through clean, reliable APIs, and build complex automated workflows using tools like n8n that connect platform products such as GraphDB, Graph Modeling, and Semantic Analytics.
- Uphold the team’s high technical bar through reviews, AI evaluation standards, and hands-on contributions, debugging complex issues across distributed AI and data pipelines.
- Diagnose and remediate LLM hallucinations, gaps, and accuracy issues.
- Architect robust prompt-engineering strategies that balance quality, latency, and cost.
- Maximize LLM efficiency through token optimization and intelligent caching.
- Design for high concurrency, fault tolerance, and comprehensive observability.
- Thrive in ambiguous, rapidly evolving technical environments.
- Balance immediate tactical execution with long-term strategic product vision.
Our Stack
- Languages: Python, Java, Node.js
- Graph and semantics: the Graphwise platform, including GraphDB (our RDF graph database), and related products such as Graph Modeling and Semantic Analytics
- LLMs and RAG: LangChain/LlamaIndex, embedding models, and GraphRAG architectures
- Search and retrieval: Elasticsearch/OpenSearch, Lucene, and various other vector store connectors
- Streaming, storage, and ingestion: S3, Kafka for data, and PostgreSQL
- Automation and delivery: n8n for workflow automation (central to this role), Jenkins for CI/CD, GitLab for version control, Kubernetes and Docker for deployment, Grafana and Prometheus for observability
What makes a good candidate
- 4+ years of experience in backend, data engineering, or software engineering, with strong proficiency in Python and/or Node.js; experience with statically typed languages such as Java is a plus but not required.
- Extensive experience with workflow automation tools such as n8n, Haystack, LangDock, Airflow, or similar, and a solid understanding of API design and the integration of external services into complex data flows.
- Demonstrated experience integrating and deploying LLM-based systems (Azure AI, AWS Bedrock, Google Vertex, HuggingFace, etc.) to production, including prompt engineering, evaluation (RAGAS, LLM-as-a-judge), and building LLM and ETL pipelines for data ingestion and indexing.
- At least 1 year of experience with advanced RAG systems (beyond vanilla RAG) in production (not tutorials or demos).
- Full professional proficiency in English (our primary working language).
Note: Hands-on knowledge graph experience is not required. If you know how to ship LLM-integrated features and data/software engineering, we can get you up to speed on graphs.
Nice-to-have-s
- Hands-on experience with graph technologies (e.g., RDF(S), GraphDB, Apache Jena, Blazegraph, or similar) and graph embeddings, or professional experience integrating knowledge graphs with LLMs.
- MLOps exposure, including CI/CD for ML, experiment tracking, model versioning, and production monitoring.
- Familiarity with multi-agent workflows and related orchestration patterns.
- Familiarity with even-driven systems (e.g., Kafka/Redis)
- Experience with distributed systems and cloud platforms (AWS, GCP, Azure).
- Hands-on experience with schema languages (RDFS/OWL, SHACL, JSON Schema, etc.)
- Contributions to relevant open-source projects.
What you will get
- The opportunity to work on cutting-edge projects in AI, knowledge graphs, and enterprise data.
- Flexible working arrangements and a healthy work-life balance.
- Competitive compensation and performance-based bonus opportunities.
- Additional health insurance and well-being benefits.
- A collaborative, international environment with colleagues across Europe, North America, and APAC.
- Access to learning resources, knowledge-sharing initiatives, and opportunities for continuous professional growth.
- The chance to work alongside some of the leading experts in semantic technologies and AI.
- Team gatherings, company events, and opportunities to connect beyond day-to-day work.
Graphwise is an equal opportunity employer. We do not and will not discriminate on the basis of age, disability, sex, race, religion or belief, gender reassignment, marriage or civil partnership status, pregnancy or maternity, or sexual orientation. All employment decisions, including recruitment, hiring, development, promotion, compensation, and other terms and conditions of employment, are based on merit, qualifications, performance, and business needs. We are dedicated to providing an environment free from discrimination, harassment, and bias, and we welcome applications from all qualified candidates.