Happy to share that our paper on new learned indexing approach for DNA sequence search - LISA - is selected for oral presentation at #NeurIPS2019 workshop on Systems for ML. Looking forward to our talk this Friday, the 13th. Tim Kraska Nesime Tatbul Jialin Ding Mohd Vasimuddin Intel AI

ALEX (our adaptive learned in-mem range index) is being presented at #SIGMOD2020 today. Up to 4X faster and 2000X smaller than B+Tree. Paper @ dl.acm.org/doi/pdf/10.114… + prototype is now OSS @ github.com/microsoft/ALEX. Comments welcome; main contacts Jialin Ding & Umar F. Minhas!

ML tool automatically reorganizes a dataset’s storage layout based on users' query types. -Runs queries 10x faster than the SotA -Uses learned indexes that are ~100x smaller than traditional indexes Paper: arxiv.org/pdf/2006.13282… More: bit.ly/MITsunami v/Tim Kraska

Our work on learned storage systems and query optimizers is featured on MIT News. Super proud on my team and collaborators, especially Jialin Ding and Ryan Marcus who lead the work news.mit.edu/2020/mit-data-…

We just launched Multidimensional Data Layouts in Amazon Redshift, a new feature that's inspired by ML for systems research from my PhD and achieved through the collaborative work of an amazing team at AWS. Paper coming soon! aws.amazon.com/blogs/big-data…

Our industry paper on "Automated Multidimensional Data Layouts in Amazon Redshift" will be presented at SIGMOD 2024! We highlight the practical challenges of deploying ML-for-databases research ideas in a large-scale commercial database product. amazon.science/publications/a…

Delighted to announce that Parachute 🪂 will appear at VLDB 2025 🇬🇧! 🇬🇧 Compared to regular semi-join filtering, Parachute removes dangling tuples in a bi-directional manner by precomputing fingerprint columns. Dangling tuples ⏬ = Join pruning ⏫. 📎 arxiv.org/abs/2506.13670

Meeting awesome people at the #edsurge summit :) Jill Bromenschenkel Jialin Ding Varun Arora #diego