Senior Engineer, Knowledge Graphs
Most retrieval for AI agents today is vector search over chunks of text. That works until a question needs several steps. Ask which trials enrolled patients on a drug made by a company that was later acquired, and it guesses.
Redpine licenses non-public data in medicine, science, law, and finance from publishers, research institutions, and other rights holders. We serve it to agents through our API, MCP, and CLI. We want those agents to follow a question through the data step by step and see where each answer came from. That takes a knowledge graph, and we’re hiring the first engineer to build it.
The role
You’ll own the knowledge graph from the start. You’ll decide what it should do and which domain to begin with, design the ontology and schema, and build the pipelines that turn raw sources into the graph. You’ll also run it in production and find out whether it actually helps agents give better answers.
Some of the problems are hard. The same entity appears under different names in different sources. Sources disagree, and the graph has to decide what to state and what to flag. Every node and edge has to trace back to the document and license it came from, because that traceability is part of what customers pay for. Sources also change, so the graph has to keep up. You’ll work with licensed data that most engineers never get access to, in fields where a wrong answer has real consequences.
What we’re looking for
You’ve built and run a knowledge graph or entity resolution system in production.
You’ve owned a system from design through production on a small team, without a platform team behind you.
Strong Python, and data pipelines you designed yourself.
A good sense of when a graph is the right tool and when simpler retrieval is enough.
Nice to have
Startup or early-team experience
RDF or OWL, or property graphs (Neo4j)
Domain ontologies such as UMLS, SNOMED CT, MeSH, or FIBO
RAG or hybrid retrieval, ideally backed by a graph
Experience with data in medicine, science, law, or finance
About Redpine
If models were the first wave of AI, and compute the second, we are building the data layer that comes next. Only a small fraction of the world's data is on the open internet. The rest, high-quality, domain-specific, often critical, sits behind paywalls, in databases, or with rights holders. Redpine is building the infrastructure to unlock it.
We provide AI builders and autonomous agents with access to licensed, high-quality, multimodal data through a unified platform and API. The goal is simple: make AI systems more accurate, more useful, and grounded in real-world information.
You will work closely with the founding team, including Anders (ex-partner at VC, McKinsey) and David (ex-tech and product leader at Spotify, Zettle, and Lunar). We are backed by Nordic Ninja, Node VC, and Luminar, alongside angels from OpenAI, Spotify, and Perplexity.
If you are eager to make a meaningful impact in the AI space, we would love to hear from you. We are committed to building a diverse and inclusive team. If you are excited about this role but your experience does not align perfectly with every qualification, we encourage you to apply anyway.