Elton Pinto

Research

I currently work on LLM inference systems and solver-based job scheduling. On the inference side, I'm a project lead on Vajra, studying hierarchical prefix caching and tensor exchange mechanisms. More broadly, I enjoy solving problems at the intersection of systems and programming languages.

Last summer I interned at Uber's Programming Systems Group, where I developed a parallel version of the RTA callgraph construction algorithm for the Go programming language.

During my master's, I worked with Daan Leijen on prototyping an implementation of Perceus memory management for OCaml.

During my undergrad I was part of TINKER lab where I was advised by Prof. Thomas Conte and Dr. Jeff Young. While there, I wrote a space-efficient implementation of the quantum verification of matrix products (QVMP) algorithm.

Fellowships

• Sutter Hill Ventures Codepoint Fellowship (2023-2024)
• NSF GRFP (declined to pursue SHV fellowship)

Publications

• RPCShield: Defending Microservices Against Cascading Failures
  Milind Chabbi, Sonal Mahajan, Ivan Beschastnikh, René Just, Yuxin Wang, Yufan Xu, Elton Pinto, Seemanta Saha, Manu Sridharan, Abhishek Jha, Sandeep Koushik Sheshadri, Mayank Bansal
  SOSP'26
  
• Revati: Transparent GPU-Free Time-Warp Emulation for LLM Serving
  Amey Agrawal, Mayank Yadav, Sukrit Kumar, Anirudha Agrawal, Garv Ghai, Souradeep Bera, Elton Pinto, Sirish Gambhira, Mohammad Adain, Kasra Sohrab, Chus Antonanzas, Alexey Tumanov
  arXiv
  
• Maya: Optimizing Deep Learning Training Workloads using GPU Runtime Emulation
  Srihas Yarlagadda*, Amey Agrawal*, Elton Pinto*, Hakesh Darapaneni, Mitali Meratwal, Shivam Mittal, Pranavi Bajjuri, Srinivas Sridharan, Alexey Tumanov
  EuroSys'26
  
• Exploring Perceus for OCaml
  Elton Pinto, Daan Leijen
  Higher-order, Typed, Inferred, Strict: ML Family Workshop, ICFP'23
  
• Perceus for OCaml
  Elton Pinto
  M.S. Thesis. Georgia Institute of Technology, May 2023.
  
• Neko: A quantum map-filter-reduce programming language
  Elton Pinto
  Student Research Competition (SRC). POPL'23
  abstract poster
• An Implementation of the Quantum Verification of Matrix Products Algorithm
  Elton Pinto, Jeffrey Young, Thomas Conte, Austin Adams, Eugene Dumitrescu
  4th International Workshop on Quantum Resource Estimation (QRE). ISCA'22
• An Implementation of the Quantum Verification of Matrix Products Algorithm
  Elton Pinto
  B.S. Thesis. Georgia Institute of Technology, May 2022.
  Third place (Explore category), UROP Symposium 2022
• Enabling a Programming Environment for an Experimental Ion Trap Quantum Testbed
  Austin Adams, Elton Pinto, Jeffrey Young, Creston Herold, Eugene Dumitrescu, Thomas Conte
  2021 International Conference on Rebooting Computing (ICRC'21)