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.
Watch the conversation
Himanshu Singh
Leads engineering on Netflix's Entertainment Knowledge Graph, having previously built and scaled knowledge graphs at eBay and on Microsoft Bing's Satori. His career has centred on one problem: turning low-quality, scattered information into cohesive, connected, high-quality knowledge. He presented at the Knowledge Graph Conference on scaling Netflix's Entertainment Knowledge Graph with LLMs.
What we'll cover
- 01Three graphs, three scales: a politics domain inside Microsoft's Satori, building from zero at eBay, and the Entertainment Knowledge Graph at Netflix.
- 02What was actually common across a general web knowledge base, an e-commerce product graph and an entertainment catalogue, once you take scale out of the comparison.
- 03Why meaning lives in the relationships rather than the node count, and how to explain that to a business that measures things by size.
- 04Federate, don't duplicate: keeping your CRM or ERP as the system of record and reaching into it, instead of copying it into the graph and inheriting the bloat.
- 05Graph databases evaluated over years and at real scale, what worked, what tapped out, and why the storage choice matters less than people assume.
- 06The reliability argument: moving the human from downstream of the output to upstream of the knowledge asset, and what that looks like operationally at Netflix's scale.
- 07Feedback loops that amount to continuous knowledge-graph auditing, and the case for an observability layer over knowledge creation rather than model output.
- 08Property graph, triple store, or neither, and whether a knowledge graph is meaningfully different from a graph database.
- 09Whether knowledge graphs changed at all after transformers, or whether only the tooling around them did.
- 10His one piece of advice for teams reaching for a graph to fix agent reliability: start from the problem and the data, not the database.
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.
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.
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.