profile_picture
David Pissarra
Ph.D. Student, New York University
david [dot] pissarra [at] nyu [dot] edu

I am a first-year Computer Science Ph.D. student at the NYU Systems Lab, fortunate to be co-advised by Prof. Aurojit Panda and Prof. Jinyang Li. My research interests lie in the intersection of Computer Systems and Machine Learning.

Prior to joining NYU, I got my Master’s degree at Instituto Superior Técnico, University of Lisbon. I worked on In-network ML inference, advised by Prof. Fernando Ramos and Prof. Muhammad Shahbaz. I was also a research intern at the Catalyst Lab working on Machine Learning Compilation for LLMs, advised by Prof. Tianqi Chen.

Research Interests

  • Machine Learning Systems
  • Compiler Optimization
  • Large-Scale Machine Learning
  • Networking

Education

New York University
2024 - Present
Ph.D. Computer Science
Instituto Superior Técnico
2022 - 2023
M.Sc. Computer Science and Engineering
Tsinghua University
2021 - 2022
Graduate Visiting Student, Advanced Computing
Instituto Superior Técnico
2018 - 2021
B.Sc. Computer Science and Engineering
Academic Merit Award

Experience

Catalyst Lab, Carnegie Mellon University, Fall 2023, Research Intern
I worked on Machine Learning Compilation for Large Language Models (MLC-LLM project). Advised by Prof. Tianqi Chen.
Instituto Superior Técnico, Fall 2021 - Spring 2024, Teaching Assistant
I was invited to teach Computer Organization (Fall 2021), Search and Planning (Fall 2023), Distributed Systems (Spring 2024), and Artificial Intelligence (Spring 2024) courses.
Fraunhofer Portugal, Summer 2023 - Spring 2024, Machine Learning Researcher
Worked on deploying and fine-tuning Large Language Models to perform textual clinical tasks.
Sensei Tech, Summer 2022, Machine Learning Researcher
Designed a Machine Learning system capable of automatically detecting annotation errors in image retail product datasets. Image Dataset Anomaly Detection.
Deloitte, Summer 2021, Software Engineer Intern
During this internship, contacted with various SAP technologies, completing an openSAP course addressing the SAP Cloud Application Programming Model (SAP CAP). Developed a small SAP CAP demo application, connecting it to multiple APIs and other SAP modules (such as SAP Fiori).

Publications

Unlocking the Potential of Large Language Models for Clinical Text Anonymization: A Comparative Study, 2024, PrivateNLP @ ACL 2024
David Pissarra , Isabel Curioso , João Alveira , Duarte Pereira , Bruno Ribeiro , Tomás Souper , Vasco Gomes , André Carreiro , Vitor Rolla
In-network ML-based Anomaly Detection, 2023, Master Thesis
David Pissarra

Selected Projects

MLC LLM, 2023, Contributor
MLC LLM is a universal solution that allows any language model to be deployed natively on a diverse set of hardware backends and native applications, plus a productive framework for everyone to further optimize model performance for their own use cases. Everything runs locally with no server support and is accelerated with local GPUs on phones and laptops.