Everything was associated with everything else
Mike Dillinger built the Economic Graph at LinkedIn, worked on knowledge graphs at eBay, and now runs hypergraf.ai, which he describes as an X-ray machine for other people's graphs. He tells the story of an organisation that spent millions building out "associated with" relations until everything in the graph was, one way or another, associated with everything else, and the query expansion got so robust it undermined the system it was built to serve. Underneath the story is a linguist's diagnosis of why these systems fail: language models operate on strings, and a string is not a concept. Query canine, Barker, Pooch and Puppy and you get four different answers. His conclusion is that the human check everyone has pushed downstream of the output belongs upstream, on the knowledge foundation.
Watch the conversation
Mike Dillinger
A linguist and cognitive scientist by training who spent his career making the case that explicit, structured knowledge is what makes AI reliable rather than merely large. He was Technical Lead for Knowledge Graphs in LinkedIn's AI division, where he built the Economic Graph powering job recommendations and the skills taxonomy at billion-user scale, and built LinkedIn's and eBay's first machine translation systems before LLMs existed. He now runs hypergraf.ai, consulting on enterprise knowledge-graph architecture, and writes a widely read newsletter on knowledge architecture.
What we'll cover
- 01From a PhD on simultaneous translation to the LinkedIn Economic Graph, and what hypergraf.ai's "X-ray machine for your knowledge graph" actually inspects.
- 02Knowledge graph, ontology, taxonomy: the three terms every AI panel now uses interchangeably, separated cleanly and without product.
- 03Why a graph is treated as a source of truth in a way a vector index is not, and where the hybrid actually lives, localise with embeddings, then traverse.
- 04Node granularity in practice: what changes downstream when your graph has one node for a car model versus separate nodes for each variant.
- 05The moment it stopped being about the model. Strings are not concepts, and no amount of scale converts one into the other.
- 06Multimodal models swap strings for pixels and audio, and the ambiguity survives the swap, what a concept space would have to look like instead.
- 07Why markdown files are not a knowledge layer, however many of them you write.
- 08Scaling from thousands of nodes to millions: what breaks, and why a sparse graph underperforms a dense one regardless of node count.
- 09Overreliance on "associated with", the relation that quietly tells you nothing, and why relations define nodes rather than the other way round.
- 10The takeaway Vivek pulled out of it: validate the knowledge foundation, not the output. Put the human upstream.
Related episodes
Bring the graph to the data
A citizen in Italy asks a public chatbot when their civil-service exam is. The honest answer keeps moving: dates get corrected, sessions get cancelled, and amendments pile up across official notices. Ask a language model alone and it answers confidently, sometimes with a date it picked at random, sometimes with a session that no longer exists. Giuseppe Futia built the knowledge graph that keeps the answer current. It does the deterministic cross-document work an LLM cannot be trusted with, matching each amendment to the exam it modifies, superseding old versions, cascading cancellations, and staying auditable throughout, inside a production pipeline behind a chatbot he says serves close to a million citizens. A former La Stampa journalist with a PhD from Politecnico di Torino, and the first European guest on the show, he also opens up the regulated side of his work: healthcare data that cannot leave the country, inference that runs on premise, and patient records he is not allowed to move even inside his own infrastructure. His portable lesson is the entry point. Prove value on one well-defined task the institution already needs, then earn the right to expand.
Relationships should be the product
Himanshu Singh has built knowledge graphs three times at three very different scales: a politics subgraph inside Microsoft's Satori, a zero-to-one product graph at eBay, and now Netflix's Entertainment Knowledge Graph, where he leads engineering. The line he keeps returning to is that relationships should be the product. Node count is not the measure, and a graph that duplicates what already lives in your CRM or your warehouse is mostly cost. He is notably relaxed about technology choice, pointing out that Netflix built its own real-time graph abstraction over a key-value store because no native graph database could absorb their write volume. What he is not relaxed about is data quality at the point of entry, because once bad data is in a graph and connected to everything else, undoing it is very hard.
A different kind of answer, not a better number
Karthik Soman invented KG-RAG — knowledge-graph-based retrieval augmented generation — while building biomedical knowledge graphs at UCSF, and now leads enterprise-scale agentic AI at SAP America. One question carries from a PhD in computational neuroscience to the enterprise: how do you build intelligent systems that actually work in the real world? His answer isn't a better accuracy number but a different kind of answer — one a human can trace, question, and learn from. He walks through the case that convinced him: enriching patient records with a 40-million-node biomedical graph surfaced an olfactory-receptor gene that flagged Parkinson's five years early, catching a prodromal case a clinician had missed — not because the model was more accurate, but because it pointed at a mechanism. Then he moves to the enterprise, where the curated ontologies of biomedicine don't exist. You lean on the topology already inside your documents. Graph-based reasoning turns out to be a sixty-year-old idea that LLMs merely made usable on the fly. And a graph earns its keep over vector RAG in specific places — multi-hop questions, smaller models, private data the model never saw — before the least glamorous advice in AI: data hygiene first, then AI hygiene.