Berlin Tech Meetup: The Future of Relational Foundation Models, Systems, and Real-World Applications
For CDOs & Chief Analytics Officers
Your models flatten relational data into feature tables, destroying the multi-hop relationships that encode the most valuable predictive signals. KumoRFM learns from that relational structure directly, capturing what traditional ML never could. On top of that, KumoRFM is pre-trained on thousands of proprietary and public relational datasets, bringing pattern knowledge your team has never seen and delivering an additional 10-50% accuracy boost (proven on Stanford RelBench). And where feature engineering takes your team months per model, KumoRFM eliminates it entirely, shipping production models in hours so the same team delivers 50+ models per quarter instead of 3-5.
Book a demo and get a free trial of the full platform: research agent, fine-tune capabilities, and forward-deployed engineer support.
By submitting, you accept the Terms and Privacy Policy.
Why Chief Data Officers choose Kumo
Here's exactly how Kumo changes the operating model of your data organization.
17x
More models shipped per quarter
One enterprise customer went from 3 models in production to 50+ in a single quarter — same team, same budget. Feature engineering eliminated entirely.
$100M+
Revenue impact at DoorDash
Restaurant recommendations driving hundreds of millions in GMV. Then expanded to notification reranking and send-time optimization — all from one platform.
95%
Reduction in pipeline complexity
No feature stores, no custom ETL per model, no data scientist bottleneck. Your team defines predictions in a query language. Kumo handles the rest.
Loved by data scientists, ML engineers & CXOs at
In production today
Enterprise Customer
Went from 3 models in production to over 50 in a single quarter with the same team. Feature engineering eliminated, pipeline complexity reduced by 95%.
Databricks
Lead-scoring models delivering dramatic improvement in conversion rates. Deployed in days instead of months, with zero feature engineering.
DoorDash
Restaurant recommendations driving hundreds of millions in GMV. Expanded to notification reranking and send-time optimization using the same foundation.
The challenge you know too well
Path 1 — Hire ML engineers: Each model costs $50K–$1M+ and ties up ~2 FTEs for months of feature engineering and pipeline work. The result: 3–5 models per year, and 53–88% never reach production. Your backlog of prediction requests keeps growing.
Path 2 — Try LLMs: They tokenize your relational data as text. They have no concept of primary keys, foreign keys, or the relationships between tables — exactly the structure that encodes the answers to your most valuable predictive questions.
KumoRFM connects directly to your data warehouse and answers these questions out of the box. Your team defines predictions in a query language. The foundation model handles everything else — churn, LTV, fraud, recommendations, all from one platform.
95%
Less data preparation
Feature engineering eliminated
10–50%
Accuracy improvement
Over traditional ML (RelBench)
20x
Faster time-to-value
From months per model to hours
55+
Use cases
Validated in production
KumoRFM isn't limited to your data alone. Pre-trained on billions of relational patterns across diverse datasets and fine-tuned to your schema, it sees what no in-house model can. As per the SAP SALT benchmark.
LLM
GPT4 + AutoML
PhD Data Scientist
Feature eng. + XGBoost
KumoRFM
Relational Foundation Model
17x
increase in models shipped per quarter
“Beating internal XGBoost model on key metrics with far less data and features. We went from three models in production to over fifty in a single quarter, with the same team.”
Matt Loskamp
GTM Data Science Leader, Enterprise Customer
From startups to enterprises, leading organizations rely on Kumo to deliver predictive insights at scale.
Peer-reviewed
Kumo is built on 40+ peer-reviewed papers at NeurIPS, ICML, and KDD. The methodology is public and reproducible.
K. Huang, M. Fey, J. Leskovec et al.
A foundation model for relational data - pre-trained across schemas, it delivers accurate predictions out of the box and improves with fine-tuning on your specific data.
Read paperM. Fey, W. Hu, K. Huang, J. Leskovec et al.
Introduces learning predictive models directly on relational databases, eliminating the feature engineering pipeline that has historically bottlenecked enterprise ML.
Read paperJ. Robinson, R. Miao, K. Huang et al.
An open benchmark for evaluating relational prediction methods across 11 databases and 30 tasks. Kumo consistently outperforms traditional ML baselines.
Read paper
