Beyond XGBoost: An Evening for ML Engineers & Leaders
Beyond XGBoost: An Evening for ML Engineers & Leaders
For over ten years, if you had a table, you reached for XGBoost.
That's changing. Every major lab now has a model in this space.
And yet most of the people running prediction in production have never heard of any of it. That gap is why we're doing this.
The conversation will explore
Whether tabular foundation models represent the next major shift in enterprise prediction, or whether a well-tuned XGBoost still wins where it matters.
What in-context and zero-shot prediction actually change.
The tradeoff between GPU cost and engineering overhead.
How these models scale across rows, columns, and dataset complexity.
Explainability, calibration, licensing, data privacy, and vendor lock-in.
Why a table is not simply another form of text.
What the prediction stack may look like over the next two to three years.
Who you'll be talking with
Join Synthefy and House of 434 for an evening with researchers, founders, applied ML leaders, investors, and technical operators exploring where these models work, where they struggle, and what it takes to deploy them at scale.
The evening opens with welcome remarks from Synthefy and Shilpi Nayak, Founder of House of 434, followed by a moderated panel of practitioners working across tabular and time series prediction.
Drinks and light bites throughout, with real time to talk before and after the panel.
The speaker lineup is coming together now. If you or someone you know should be on it, tell us when you request an invitation.
Building in this space?
If you are building, researching, or deploying tabular or time series foundation models, tell us about your work when requesting an invitation.
A small number of founders, researchers, and applied ML leaders may be invited to participate in the panel or contribute to the conversation.
The largest applications for foundation models include: supply chain, predictive maintenance, fraud and churn detection, performance marketing.
Agenda:
5:30 PM · Doors open, drinks and networking
6:00 PM · Welcome remarks
6:15 to 6:45 PM · Moderated panel on foundation models for structured data, how they compare with traditional machine learning approaches, and what it takes to deploy them in production
6:45 PM · Drinks, light bites, and open networking
8:00 PM · Event concludes
About Synthefy
Synthefy is building foundation models for structured and time series data, helping organizations approach prediction problems with less feature engineering, model tuning, and dataset-specific development.
Interested in learning more? Check out our hugging face where you can test your own predictions in real-time:
https://huggingface.co/Synthefy/Nori
About House of 434
House of 434 is a platform connecting founders, investors, operators, and domain experts through curated gatherings, strategic conversations, and meaningful introductions. We bring together the people shaping the future of technology, innovation, and private markets.
https://houseof434.com/