Berlin Tech Meetup: The Future of Relational Foundation Models, Systems, and Real-World Applications
For Heads of Ads & Ad Tech
Lookalike audiences rely on flat demographic features, missing the relational graph of behavior, purchases, and interactions. KumoRFM understands that full graph natively. It's also pre-trained on thousands of relational datasets, providing additional pattern knowledge that pushes targeting accuracy far beyond what your current models can achieve. And it delivers production-grade scoring in hours, not the months your current pipeline takes.
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 Ad Leaders choose Kumo
Here's exactly how Kumo changes ad performance for your team.
10–50%
More accurate targeting
KumoRFM understands the full relational graph — user behavior, content interactions, purchase patterns, social connections — for targeting that goes far beyond demographic lookalikes.
Real-time
Propensity scoring
Score every user's likelihood to convert, subscribe, or churn in real-time via API. Update targeting models continuously as new data flows in.
5.4x
Conversion rate improvement
Databricks saw 5.4x improvement in lead-scoring accuracy. Apply the same approach to ad conversion prediction and bid optimization.
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 — Lookalike audiences: They use flat feature vectors, collapsing the rich graph of user-product-content relationships into demographic buckets. You lose the signal that actually predicts intent — what users browse, who they interact with, and how those patterns evolve over time.
Path 2 — LLMs for ad scoring: They tokenize your relational data as text. At real-time ad scoring scale — millions of bid decisions per second — LLMs are too expensive and too slow. They also have no concept of the relational structure that encodes your most valuable signals.
KumoRFM delivers production-grade predictions that understand relational patterns — user-product affinities, content engagement graphs, purchase sequences — and scores them in real-time via API. Better targeting, smarter bids, higher ROAS.
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
