Andreas Kipf

Professor @utn_nuremberg. Formerly @AWS, @MIT_CSAIL, @TU_Muenchen. Focused on building efficient, easy-to-use data systems.

[Andreas Kipf @andreaskipf]

Excited to share that I’ll start as Professor of Data Systems @utn_nuremberg in early 2024! My research will explore the intersection of data systems and ML.

I’ll soon announce PhD and postdoc positions in my group.

[ACM SIGMOD/PODS (@sigmodconf.bsky.social) @SIGMODConf]

📣 🏆2026 SIGMOD Research Highlight Awards 🏆:
https://t.co/r5jftg2zNz

DPconv: Super-Polynomially Faster Join Ordering
@mihail_sto, @andreaskipf
https://t.co/f38OeqGoWr

Automating Vectorized Distributed Graph Computation
Wenyue Zhao, Yang Cao, Peter Buneman, Jia Li, @ntarmos
https://t.co/4BL6HB16o7

AnyBlox: A Framework for Self-Decoding Datasets
Mateusz Gienieczko, @maxikuschewski, Thomas Neumann, Viktor Leis, @JanaGiceva
https://t.co/qxqdMMjCw3

Rel: A Programming Language for Relational Data
@molhamaref, Paolo Guagliardo, @gkastrinis, Leonid Libkin, Victor Marsault, Wim Martens, Mary McGrath, Filip Murlak, Nathaniel Nystrom, Liat Peterfreund, Allison Rogers, Cristina Sirangelo, Domagoj Vrgoč, David Zhao, Abdul Zreika
https://t.co/xO5v44v4Yi

MEMPHIS: Holistic Lineage-based Reuse and Memory Management for Multi-backend ML Systems
@ArnabPhani, @matthiasboehm7
https://t.co/YRbr8x1nov

Diva: Dynamic Range Filter for Var-Length Keys and Queries
Navid Eslami, @IoanaBercea, @niv_dayan
https://t.co/ddwHqGHeaC

The Key to Effective UDF Optimization: Before Inlining, First Perform Outlining
Samuel Arch, Yuchen Liu, Todd C. Mowry, @pateljm, @andy_pavlo
https://t.co/yclqSE6VFT

Output-sensitive Conjunctive Query Evaluation
@ShaleenDeep, @HangdongZ79542, Austen Z. Fan, Paraschos Koutris
https://t.co/5aj48xeUZJ

Output-Optimal Algorithms for Join-Aggregate Queries
Xiao Hu
https://t.co/VbrUbnXarQ

Differentially Private Substring and Document Counting
Giulia Bernardini, @philipbille, @li_rtz, Teresa Anna Steiner
https://t.co/GYCjSDeV9l

Congratulations to all the authors👏 👏 💐
#SIGMOD2026 #ACM #researchhighlight #SIGMODawards

[Mihail Stoian @mihail_sto]

Can your cloud database predict underprovisioning before it even happens?

Meet ◒ xBound, the very first framework for join size lower bounds. xBound tells you how many tuples your SQL query will produce *at least*.

Brought to you by Microsoft GSL & @utndatasystems.

[Andreas Kipf @andreaskipf]

I hope you've had a great start to the year! I'm excited to announce our blog. We're kicking things off with a look back at everything that happened in 2025.

https://t.co/cU8qhHe8xt

[Mihail Stoian @mihail_sto]

PoC customer bringing their own workload to test?

👉 Boost their LIKE/REGEX predicates with 🌰 string fingerprints.

Freshly presented at AIDB'25 @VLDBconf.

Paper: https://t.co/1Hjg7zniPq
Code: https://t.co/86NG2JFz9W

[Andreas Kipf @andreaskipf]

Looking forward to great discussions and catching up with everyone at VLDB!

[Andreas Kipf @andreaskipf]

Our lab is excited to be presenting two papers at VLDB 2025 in London this week! 🇬🇧

[Andreas Kipf @andreaskipf]

Tuesday, 1:45 PM:
🪂 Parachute: Single-Pass Bi-Directional Information Passing
by Mihail Stoian (Research 8 — Westminster, 4F)
PDF: https://t.co/1yXKqm33hR
Code: https://t.co/Ih90BXxbsn

[Yuki @y_m_asano]

Today we release Franca, a new vision Foundation Model that matches and sometimes outperforms DINOv2.
The data, the training code and the model weights (with intermediate checkpoints) are open-source, allowing everyone to build on this.

Methodologically, we introduce two new SSL components, one is a multi-granularity SK clustering loss that utilizes Matryoshka representations and a quick post-pretraining scheme to remove unwanted spatial biases.

This is the result of a close and fun collaboration @valeoai (in France) and @FunAILab (in Franconia)

[Andreas Kipf @andreaskipf]

🔗 Learn more about our research and team: https://t.co/HpsGiaSuwR

[Andreas Kipf @andreaskipf]

The Data Systems Lab is seeking a motivated PhD candidate to join our team and work on foundation models for data compression.

[Andreas Kipf @andreaskipf]

🛠️ The position requires strong programming skills in C++ and Python.

We've already published early results in this space:
- Virtual, TRL @ NeurIPS'24: https://t.co/rfiQutLm3v
- Virtual, EDBT'25 (Best Demo): https://t.co/4rcgVFC7yu

[Andreas Kipf @andreaskipf]

Off to SIGMOD 2025 in Berlin! 🚄

Here’s our schedule:

Today, 4:20 PM:
💡 Redbench: A Benchmark Reflecting Real Workloads (aiDM)

Wed, 2:00 PM:
🏆 DPconv: Super-Polynomially Faster Join Ordering

Thu, 2:30 PM:
❄️ Pruning in Snowflake: Working Smarter, Not Harder

Come say hi! 👋

[Andreas Kipf @andreaskipf]

Parachute takes semi-join filtering to the next level!

Congrats to my PhD student @mihail_sto and thanks to our co-authors from MIT for initiating the project four years ago.

See you in London! 🇬🇧

[Mihail Stoian @mihail_sto]

Delighted to announce that Parachute 🪂 will appear at @VLDBconf! 🇬🇧

Compared to regular semi-join filtering, Parachute removes dangling tuples in a bi-directional manner by precomputing fingerprint columns.

Dangling tuples ⏬ = Join pruning ⏫.

📎 https://t.co/P80WXLb6Qp

[Andreas Kipf @andreaskipf]

Fantastic news 🎖️

@mihail_sto will present DPconv at SIGMOD in Berlin this June.

[Mihail Stoian @mihail_sto]

DPconv just won a SIGMOD'25 Honorable Mention! 🥁

I was quite impressed given this year's high-quality papers. Let's see who won the big prize.

My list of candidates in the thread below 🧵.

[Andreas Kipf @andreaskipf]

Thrilled to share that we've received the Best Demonstration Award 🏆 at EDBT 2025!

Congratulations to my students @mihail_sto and Ping-Lin Kuo for their excellent work and dedication over the past few weeks—well deserved!

Paper: https://t.co/4rcgVFBzIW

[Mihail Stoian @mihail_sto]

Check out our poster tomorrow at EDBT Demo 🇪🇸!

🔥Update: Virtual v0.2 now supports S3 and 🤗 Parquet files - try it out!

pip install virtual-parquet.

[Andreas Kipf @andreaskipf]

We just released Redbench, a new benchmark that contains 30 analytical SQL workloads that can be used to benchmark workload-driven optimizations. Go check it out!

GitHub: https://t.co/LKBsXc8347