Berlin Tech Meetup: The Future of Relational Foundation Models, Systems, and Real-World Applications
For Senior Staff ML Engineers
The build-vs-buy math isn't just about cost. It's about accuracy. Your hand-built models flatten relational data, missing multi-hop signals. KumoRFM learns from that structure directly. It's also pre-trained on thousands of datasets your team has never seen, providing additional pattern knowledge that pushes accuracy 10-50% beyond XGBoost/LightGBM baselines on RelBench. And it compresses months of engineering into hours.
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 Senior Staff MLEs choose Kumo
You can evaluate every claim below against public benchmarks and your own data. No black-box vendor promises.
$50K–$1M
Cost per hand-built model
Each custom model ties up 2+ FTEs for months: feature engineering, pipeline development, deployment, monitoring. Kumo reduces the marginal cost of each new model to near zero.
RelBench
Open benchmark you can verify
KumoRFM is evaluated on Stanford's RelBench — 11 databases, 30 tasks. Results are public and reproducible. No black-box vendor claims.
Native
Snowflake & Databricks integration
Runs as a native app in your existing data platform. No data movement, no new infrastructure. Evaluate it in your environment with your data.
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 real question
Path 1 — Build custom pipelines: Your team can build custom ML pipelines, but each model consumes months of senior engineering time that could be spent on differentiated problems. Feature stores, training infrastructure, serving layers — you maintain all of it.
Path 2 — Try LLMs: LLMs are impressive for text but fundamentally the wrong architecture for structured relational prediction. They tokenize your tables as text, discarding the primary keys, foreign keys, and multi-hop relationships that encode your most valuable signals.
KumoRFM is purpose-built for relational databases — open methodology, benchmark-verified on Stanford's RelBench, and runs as a native app in your existing Snowflake or Databricks environment. No data movement. No new infrastructure. Evaluate it with your own data.
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
