Matteo Interlandi


I am a Principal Research Scientist in GSL at Microsoft, working at the intersection between Machine Learning and Database systems. My work has received an honorable mention at SIGMOD 2021, a best demo award at VLDB 2022, and it was featured in the “Best of VLDB 2016”.

Before Microsoft, I was a Postdoctoral Scholar in the CS Department at the University of California, Los Angeles under the supervision of Professor Tyson Condie and working on Big Data systems. Prior to joining UCLA, I was Research Associate at the Qatar Computing Research Institute and at the Institute for Human and Machine Cognition. I obtained my PhD in Computer Science at the University of Modena and Reggio Emilia.


News

  • 08/05/2023 - Our paper on A Deep Dive into Common Open Formats for Analytical DBMSs won the Best Paper Runner-Up Award for the Experiment, Analysis, & Benchmark Track @ VLDB 2023!

  • 07/15/2023 - Our paper on A Deep Dive into Common Open Formats for Analytical DBMSs has been accepted @ VLDB 2023!

  • 06/01/2023 - Our paper on Query Procesing on Gaming Consoles has been accepted to the a href="https://www.damon-db.org/home">DAMON workshop @ SIGMOD 2023!

  • 04/10/2023 - Our tutorial on Optimizing Tensor Computations: From Applications to Compilation and Runtime Techniques together with Matthias Boehm and Chris Jermaine has been accepted @ SIGMOD 2023!

  • 04/08/2023 - This year I will be serving as Sponsorship Chair @ Socc.

  • 04/07/2023 - I was one of the keynote speaker and panelist at the DBML workshop @ ICDE 2023!

  • 15/01/2023 - Our demo Unshackling Database Benchmarking from Synthetic Workloads is accept @ ICDE 2023!

  • 10/28/2022 - Our paper The Tensor Data Platform: Towards an AI-centric Database System is accept @ CIDR 2023!

  • 09/06/2022 - We won the the best demo award @ VLDB 2022 with Share the Tensor Tea: How Databases can Leverage the Machine Learning Ecosystem.

  • 09/01/2022 - This year we will be presenting @ VLDB one research paper, one demo paper, and one SDS paper. The first two works are on using PyTorch for query processing, while the last one is on making traditional ML pipelines differentiable.

  • 07/01/2022 - Our journal paper Data Science Through the Looking Glass: Analysis of Millions of GitHub Notebooks and ML.NET Pipelines is accepted @ SIGMOD Record.

  • 06/01/2022 - We got two papers accepted @ SIGMOD 2022: One on our experience on running a steering query optimized in production in SCOPE, and one co-optimized ML and SQL queries.

  • 09/01/2021 - This year we will be presenting @ VLDB one research paper, one vision paper, and one tutorial.

  • 06/20/2021 - Our paper Steering Query Optimizers: A Practical Take on Big DataWorkloads got Honorable Mention for the Industry Track @ SIGMOD 2021!

  • 08/18/2020 - Our paper Steering Query Optimizers: A Practical Take on Big DataWorkloads is accepted @ SIGMOD 2021!

  • 08/18/2020 - Our paper on Hummingbird ("A Tensor-based Approach for One-size-fits-all ML Prediction Serving") is accepted @ OSDI 2020! The technical report can be found here.

  • 05/27/2020 - We recently open sourced Hummingbird.

  • 05/15/2020 - Our paper on Building Continuous Integration Services for Machine Learning is accepted @ KDD Applied Data Science Track as oral presentation (only 44\756 got oral this year)!

  • 05/02/2020 - Our paper on Explaining data with descriptions is accepted @ Information Systems.

Previous News

Professional activities

Program Committee: ICDE 2024, EuroSys 2024, DEEM 2023, TaPP 2023, TLR Workshop, VLDB 2023, EuroSys 2022, TaPP 2022, ICDE 2022, VLDB 2021, EuroSys 2020, TaPP 2020, DASFAA 2020, DASFAA 2019, MLOSS 2018, TaPP 2018, SoCC 2018, SIGMOD 2017 (Demo Track), HotCloud 2016.

External Reviewer: VLDBJ 2023, VLDBJ 2022, CIKM 2018, SIGMOD 2017, VLDB 2016, SIGMOD 2016, JDIQ 2016, TODS 2015, VLDB 2015, ICDE 2015, VLDB 2014, VLDB 2013, ICDE 2013.


Open Source

  • 2014-2017:Titian: fine-grained data provenance in Apache Spark.

  • 2018-2020: TorchSharp: C# bindingins for Torch.

  • 2020-ongoing: Hummingbird: a tensor compiler for classical machine learning models.


Papers

  • The Tensor Data Platform: Towards an AI-centric Database System
    Apurva Gandhi, Yuki Asada, Victor Fu, Advitya Gemawat, Lihao Zhang, Rathijit Sen, Carlo Curino, Jesús Camacho-Rodríguez, Matteo Interlandi
    CIDR 2023

  • Share the Tensor Tea: How Databases can Leverage the Machine Learning Ecosystem
    Yuki Asada, Victor Fu, Apurva Gandhi, Advitya Gemawat, Lihao Zhang, Vivek Gupta, Ehi Nosakhare, Dalitso Banda, Rathijit Sen, Matteo Interlandi
    VLDB 2022

  • Query Processing on Tensor Computation Runtimes
    Dong He, Supun C Nakandala, Dalitso Banda, Rathijit Sen, Karla Saur, Kwanghyun Park, Carlo Curino, Jesús Camacho-Rodríguez, Konstantinos Karanasos, Matteo Interlandi
    VLDB 2022

  • WindTunnel: Towards Differentiable ML Pipelines Beyond a Single Model
    Gyeong-In Yu, Saeed Amizadeh, Sehoon Kim, Artidoro Pagnoni, Ce Zhang, Byung-Gon Chun, Markus Weimer, Matteo Interlandi
    VLDB 2022

  • Data Science Through the Looking Glass: Analysis of Millions of GitHub Notebooks and ML.NET Pipelines
    Fotis Psallidas, Yiwen Zhu, Bojan Karlas, Jordan Henkel, Matteo Interlandi, Subru Krishnan, Brian Kroth, K. Venkatesh Emani, Wentao Wu, Ce Zhang, Markus Weimer, Avrilia Floratou, Carlo Curino, Konstantinos Karanasos
    SIGMOD Record

  • End-to-end Optimization of Machine Learning Prediction Queries
    Kwanghyun Park, Karla Saur, Dalitso Banda, Rathijit Sen, Matteo Interlandi, Konstantinos Karanasos
    SIGMOD 2022

  • Deploying a Steered Query Optimizer in Production at Microsoft
    Wangda Zhang, Matteo Interlandi, Paul Mineiro, Shi Qiao, Nasim Ghazanfari, Karlen Lie, Marc T. Friedman, Rafah Hosn, Hiren Patel, Alekh Jindal
    SIGMOD 2022

  • Tensors: An abstraction for general data processing
    Dimitrios Koutsoukos, Supun Nakandala, Konstantinos Karanasos, Karla Saur, Gustavo Alonso, Matteo Interlandi
    VLDB 2021

  • Phoebe: A Learning-based Checkpoint Optimizer
    Yiwen Zhu, Matteo Interlandi, Abhishek Roy, Krishnadhan Das, Hiren Patel, Malay Bag, Hitesh Sharma, Alekh Jindal
    VLDB 2021

  • Steering Query Optimizers: A Practical Take on Big Data Workloads
    Parimarjan Negi, Matteo Interlandi, Ryan Marcus, Mohammad Alizadeh, Tim Kraska, Marc T. Friedman, Alekh Jindal
    SIGMOD 2021

  • Transforming ML Predictive Pipelines into SQL with MASQ
    Francesco Del Buono, Matteo Paganelli, Paolo Sottovia, Matteo Interlandi, Francesco Guerra
    SIGMOD 2021

  • A Tensor Compiler for Unified Machine Learning Prediction Serving
    Supun Nakandala, Karla Saur, Gyeong-In Yu, Konstantinos Karanasos, Carlo Curino, Markus Weimer, Matteo Interlandi
    OSDI 2020

  • Building Continuous Integration Services for Machine Learning
    Bojan Karlaš, Matteo Interlandi, Cedric Renggli, Wentao Wu, Ce Zhang, Deepak Mukunthu Iyappan Babu, Jordan Edwards, Chris Lauren, Andy Xu, Markus Weimer
    KDD 2020

  • Extending Relational Query Processing with ML Inference
    Konstantinos Karanasos, Matteo Interlandi, Doris Xin, Fotis Psallidas, Rathijit Sen, Kwanghyun Park, Ivan Popivanov, Supun Nakandala, Subru Krishnan, Markus Weimer, Yuan Yu, Raghu Ramakrishnan, Carlo Curino
    CIDR 2020

  • Cloudy with high chance of DBMS: A 10-year prediction for Enterprise-Grade ML
    The GSL Team
    CIDR 2020

Full papers list

Recent talks

  • DBML 2023, Anaheim, USA, April 2023
    How Databases and Machine Learning Systems Can Benefit from Each Other: A Perspective from Product and Research

  • HPTS 2022, Asilomar, USA, October 2022
    How to kill two birds with one stone

  • VLDB 2022, Sydney, Australia, September 2022
    Query Processing on Tensor Computation Runtimes

  • Data+AI Summit, Remote, May 2021
    Tensors Are All You Need: Faster Inference with Hummingbird

  • 15-884: Machine Learning Systems @ CMU, Remote, April 2021
    Hummingbird: A Tensor Compiler for Unified Machine Learning Prediction Serving

  • PyCon Belarus, Remote, March 2021
    Hummingbird: A Tensor Compiler for Unified Machine Learning Prediction Serving

  • COMPASS @ ETH, Remote, Febraury 2021
    Hummingbird: A Tensor Compiler for Unified Machine Learning Prediction Serving

  • Apache TVM and Deep Learning Compilation Conference, Seattle, WA, USA December 2019
    Compiling Classical ML Pipelines into Tensor Computations for One-size-fits-all Prediction Serving

All talks


Contact information