Build an AI-powered digital blueprint for enterprises using graph technologies.
Help organizations untangle complexity by building the AI-powered digital blueprint of the modern enterprise.
Ardoq is one of Norway’s most exciting scale-ups—a truly global company built out of Oslo. We are a SaaS platform that helps organizations understand, manage, and evolve their complex digital landscape. By providing a dynamic, collaborative "digital twin" of their business, we connect systems, people, and processes to drive better decision-making and accelerate digital transformation.
Today, we help some of the world’s most complex organizations (like ExxonMobil and British Telecom) gain clarity in a changing world. We are backed by leading global tech investors, including EQT and One Peak, giving us the stability of a mature company with the speed and soul of a startup.
In an AI-fueled world, enterprises face a new kind of complexity—not just more data, but more noise. As a Senior Engineer in the Graph team, you will be at the heart of our Knowledge Graph and AI-driven transformation, building the tools that turn fragmented inputs into connected, contextual intelligence.
Ardoq has been developing its own graph database engine. A unique multi-model graph database with support for point in time queries, zero cost branching, semantic models and reasoning. This technology is essential for achieving the strategic goal of developing enterprise digital twins.
At Ardoq, we value autonomy and psychological safety. In this role, you will have the influence to shape the domain and the ownership to see your ideas through to execution.
Build the engine: Extend GraphLake's storage layer, query planner and execution engine. Own changes end-to-end — data structure choice, on-disk format, cardinality estimation, traversal operators, production rollout. Work in the OneGraph model where RDF triples and labelled property graphs share storage and execution.
Make temporal and branched graphs cheap at scale: Push our point-in-time and zero-cost branching primitives — built on immutable, append-only storage with structural sharing — to the largest customer graphs without sacrificing read latency or write throughput.
Strengthen the semantic reasoning layer: Improve how the engine handles OWL ontologies, SPARQL query planning, SHACL validation, and rule-based inference.
Harden it for production: Profile, benchmark and tune the hot path. Reduce tail latency, eliminate lock contention, improve memory locality, and pay down architectural debt before it becomes a customer incident. Build the test harness that catches regressions before they ship.
You'll start by digging into a benchmark regression, a SPARQL planner edge case, or a concurrency bug surfaced overnight by load tests. You'll spend hours in the engine code — likely on meta-data files,, the persistence layer, or the inference loop — pair-debugging at the data structure level, reading flamegraphs, and reasoning about correctness under concurrent writes and time-travel reads. You'll review a colleague's design for a new traversal operator or a delta-layer encoding with the same rigour you'd want on your own. In the afternoon, you'll trade notes with Product on which semantic primitives the AI layer needs to expose, then ship the change behind a feature flag with a benchmark and a correctness proof that holds up.
We are looking for a collaborator who balances deep domain expertise with a humble, "team-first" mindset.
The Experience: You have built non-trivial systems software that ran in production — a graph database, RDF triple store, query engine, distributed database, or comparable infrastructure. You've worked on or with systems such as Neo4j, Memgraph, TigerGraph, JanusGraph, Dgraph, Stardog, RDFox, GraphDB, AllegroGraph, Virtuoso, Apache Jena, TerminusDB, Datomic, or XTDB — and you can talk credibly about the design trade-offs each one made.
The Stack: You are fluent in the semantic web stack — RDF, OWL, SPARQL, SHACL, named graphs, reification — and in at least one property-graph stack (Cypher, Gremlin, openCypher). You understand why these two worlds exist separately, where each one wins, and what's hard about unifying them. Familiarity with Datalog and description-logic reasoning is a strong plus.
The Fundamentals: You can reason about the primitives that make a database fast and correct: B-trees vs. LSM, MVCC and snapshot isolation, write-ahead logging, cost-based query optimisation and cardinality estimation, index design for join-heavy workloads, persistent (immutable, structurally shared) data structures, lock-free concurrency, memory-mapped I/O, and the difference between a cache miss and a lock contention pattern in a flamegraph.
The Engineer: You profile before you optimise, benchmark before you ship, and write the test that would have caught the bug. You read papers (the original Datomic, TerminusDB, Spanner, Dynamo, and graph-DB literature) and bring ideas back into code. You raise technical concerns early, challenge architectural assumptions respectfully, and care about long-term codebase health as much as the next ticket.
We live by three values — Bold, Caring, and Driven — that define how we work, how we treat each other, and how we win.
BOLD: We continuously raise the bar for ourselves and each other, starting from a place of "what if I could?" We act decisively, even when the information isn't complete, and choose forward motion and learning over waiting for the perfect answer.
CARING: We care enough to be candid. That means delivering direct, kind, and timely feedback that helps our teammates grow. We operate as one team — sharing knowledge openly, celebrating each other's wins, and putting collective success ahead of individual ones.
DRIVEN: At Ardoq, everyone has a stake in the outcome — including stock options. We act like owners: measuring success by impact rather than effort, saying "no" to distractions, and staying focused on the highest-value work at pace.
We believe that high performance is powered by wellbeing. You’ll have access to a competitive package that supports performance, wellbeing, and long-term success.
Employee Stock Options - share in Ardoq’s success as we grow together
25 days annual leave offered globally
Enhanced parental leave available globally to support you and your family
Retirement and insurance benefits, including travel, health, disability and life insurance
Annual learning budget to support your growth and development
A hybrid working policy: 2-3 days per week from one of our centrally located offices (London, Copenhagen, Oslo, New York)
Send us your CV - we’d love to hear from you
Our hiring process typically includes 3 - 4 interview stages:
Intro screen → chat with hiring manager → technical interview → values interview.
At Ardoq, every teammate has a voice. Whether you love tackling enterprise-scale challenges or improving how companies run, you’ll not only belong, you’ll impact, grow, and make work meaningful.
#LI-Hybrid