Berlin Tech Meetup: The Future of Relational Foundation Models, Systems, and Real-World Applications
For VPs of Data Science, ML & AI
Your team's models flatten relational data, losing the multi-hop signals that drive accuracy. On top of that, they haven't been pre-trained on thousands of relational datasets. KumoRFM has, giving it pattern knowledge your team has never seen and an additional 10-50% accuracy boost on RelBench. And each model takes months of feature engineering. KumoRFM eliminates all three problems: better accuracy from relational structure, additional gains from pre-training, and hours to production instead of months.
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 VPs of Data Science choose Kumo
Your team is talented. Your roadmap is ambitious. Here's how Kumo removes the bottleneck.
17x
More models per quarter
One team went from 3 models in production to 50+ in a single quarter. Same people, same budget — just a fundamentally different approach to building predictions.
Zero
Feature engineering required
Your data scientists stop writing SQL joins and aggregations. They write prediction queries in Kumo’s PQL. The foundation model handles feature discovery automatically.
5.4x
Conversion rate improvement
Databricks deployed Kumo-powered lead scoring in days, not months. Dramatic accuracy improvement over their previous approach — with zero feature engineering.
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 — Pipeline and infrastructure overhead: Your data scientists spend 80% of their time on feature engineering, data wrangling, and pipeline maintenance instead of building models. Each new use case means another ETL job, another feature store sync, another month before stakeholders see results. Meanwhile, the best talent on your team is burning out on plumbing work.
Path 2 — LLM limitations: Large language models flatten your relational data into text sequences. They cannot reason over primary keys, foreign keys, or multi-hop relationships across tables — exactly the structure that makes your enterprise data uniquely valuable for predictive tasks.
KumoRFM connects directly to your data warehouse and learns from the relational structure itself. Your team defines predictions in a simple query language. The foundation model handles feature engineering, training, and deployment — so your team focuses on solving business problems, not building infrastructure.
95%
Less data preparation
Feature engineering eliminated
10x
More models per team
Same headcount, 10x the output
Hours
Not months
From idea to production model
50+
Use cases validated
Across industries 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
Built on 40+ peer-reviewed papers at NeurIPS, ICML, and KDD. The methodology is public and reproducible — your team can validate every claim.
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
