Berlin Tech Meetup: The Future of Relational Foundation Models, Systems, and Real-World Applications
For ML Engineers
Your pipelines flatten relational data, capping accuracy. KumoRFM learns from the full relational graph. It's also pre-trained on thousands of datasets your pipelines have never seen, providing additional pattern knowledge that delivers 10-50% higher accuracy on RelBench. And it replaces months of custom ETL, feature stores, and serving infrastructure with hours to production.
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 ML Engineers choose Kumo
Stop maintaining custom ETL, feature stores, and serving infrastructure for every model.
95%
Less pipeline code
No custom ETL per model. No feature store. No per-model serving infrastructure. Kumo connects to your existing data warehouse and handles the entire lifecycle.
Zero
Train-serve skew
KumoRFM computes features at inference time directly from your relational data. No feature store drift. No offline/online inconsistency.
Continuous
Retraining built in
Models automatically retrain as new data arrives. No cron jobs, no data drift monitoring, no manual retraining pipelines.
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 — Custom pipelines per model: Every new use case means a dedicated ETL pipeline, feature store updates, schema drift fixes, and deployment orchestration. You spend 80% of your time on infrastructure and 20% on actual modeling. The pipeline code dwarfs the model code.
Path 2 — Try LLMs: They require entirely new serving infrastructure — GPU clusters, prompt management, token budgets, and latency optimization. And they still can't reason over the relational structure in your data warehouse.
Kumo connects directly to your data warehouse and handles training, deployment, and monitoring from a single platform. No feature stores. No custom ETL. Define predictions in a query language and the foundation model handles everything else.
95%
Less pipeline code
Custom ETL eliminated
Zero
Feature store maintenance
No more schema drift fixes
Hours
To deploy
From idea to production
50+
Models on one platform
Same infrastructure, any use case
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
