Mike Heddes

Mike Heddes

Research Software Engineer

I am Mike Heddes, a Research Software Engineer at Google. My research covers artificial intelligence (AI) and algorithms for large-scale data, ranging from model architectures and multi-agent systems to randomized sketching algorithms.

I earned my Ph.D. in Computer Science at the University of California, Irvine and my B.S. in Engineering from the Amsterdam University of Applied Sciences. I perviously co-founded the data analytics startup Garmite and conducted AI research at Google Research, Intel Labs, and the European Space Agency. I am mesmerized by all meticulous design and engineering, enjoy tackling difficult problems, and aspire to contribute to the exciting achievements of humanity. In my spare time, I enjoy making music and reading about science, business, and technology.

Google ScholarGoogle ScholarGitHub iconGitHubLinkedInLinkedIn

Experience

  • Google logo
    GoogleResearch Software Engineer
    2026 – Present
  • G
    GarmiteCo-Founder & CEO
    2024 – 2025
  • Intel LabsAI Research Intern
    2022
  • European Space Agency logo
    European Space AgencyMachine Learning Intern
    2019

Education

  • UC IrvinePh.D. in Computer Science
    2020 – 2026
  • Amsterdam University of Applied Sciences logo
    Amsterdam UASB.S. in Engineering
    2015 – 2019

Publications

DuoDiscover: Scaling Autonomous Discovery Beyond the Evaluation Bottleneck

Shubham Agarwal, Akshat Gupta, Mike Heddes, Dilara Soylu, Rui Meng, Patrick Tser Jern Kon, Jiefeng Chen, Mert Cemri, Shu Liu, Abe Gonzalez, Akanksha Jain, Bhavana Dalvi Mishra, Linfeng Song, Vilobh Meshram, Xiangru Tang, Jinsung Yoon, Aditya Parameswaran, Ion Stoica, Vahab Mirrokni, Burak Gokturk, Parthasarathy Ranganathan, Tomas Pfister, Nanyun Peng, Chun-Liang Li

Preprint ⟶

Approximating Tensor Network Contraction with Sketches

Mike Heddes, Igor Nunes, Tony Givargis, Alex Nicolau

Preprint ⟶

Approximation Algorithms for Similarity Measures and Tensor Contractions

University of California, Irvine, 2026

Mike Heddes

Dissertation ⟶

DeepCrossAttention: Supercharging Transformer Residual Connections

International Conference on Machine Learning (ICML), 2025

Mike Heddes, Adel Javanmard, Kyriakos Axiotis, Gang Fu, MohammadHossein Bateni, Vahab Mirrokni

Publication ⟶

Always-Sparse Training by Growing Connections with Guided Stochastic Exploration

International Joint Conference on Neural Networks (IJCNN), 2025

Mike Heddes, Narayan Srinivasa, Tony Givargis, Alexandru Nicolau

Publication ⟶Code ⟶

Hyperdimensional Computing: A Framework for Stochastic Computation and Symbolic AI

Journal of Big Data, 2024

Mike Heddes, Igor Nunes, Tony Givargis, Alexandru Nicolau, Alex Veidenbaum

Publication ⟶

Convolution and Cross-Correlation of Count Sketches Enables Fast Cardinality Estimation of Multi-Join Queries

International Conference on Management of Data (SIGMOD), 2024

Mike Heddes, Igor Nunes, Tony Givargis, Alex Nicolau

Publication ⟶Code ⟶

DotHash: Estimating Set Similarity Metrics for Link Prediction and Document Deduplication

Conference on Knowledge Discovery and Data Mining (KDD), 2023

Igor Nunes, Mike Heddes, Pere Vergés, Danny Abraham, Alexander Veidenbaum, Alexandru Nicolau, Tony Givargis

Publication ⟶Code ⟶

Torchhd: An Open Source Python Library to Support Research on Hyperdimensional Computing and Vector Symbolic Architectures

Journal of Machine Learning Research (JMLR), 2023

Mike Heddes, Igor Nunes, Pere Vergés, Denis Kleyko, Danny Abraham, Tony Givargis, Alexandru Nicolau, Alexander Veidenbaum

Publication ⟶Code ⟶

An Extension to Basis-Hypervectors for Learning from Circular Data in Hyperdimensional Computing

Design Automation Conference (DAC), 2023

Igor Nunes, Mike Heddes, Tony Givargis, Alexandru Nicolau

Publication ⟶

Hyperdimensional Hashing: An Efficient and Robust Dynamic Hash Table

Design Automation Conference (DAC), 2022

Mike Heddes, Igor Nunes, Tony Givargis, Alexandru Nicolau, Alex Veidenbaum

Publication ⟶
Best Paper Award Nomination

GraphHD: Efficient Graph Classification using Hyperdimensional Computing

Design, Automation and Test in Europe Conference (DATE), 2022

Igor Nunes, Mike Heddes, Tony Givargis, Alexandru Nicolau, Alex Veidenbaum

Publication ⟶

EdgeAvatar: An Edge Computing System for Building Virtual Beings

Electronics, 2021

Neftali Watkinson, Fedor Zaitsev, Aniket Shivam, Michael Demirev, Mike Heddes, Tony Givargis, Alexandru Nicolau, Alexander Veidenbaum

Publication ⟶Application ⟶

Projects