Sarath Shekkizhar, Ph.D.

Staff Research Scientist at Salesforce

Education

Ph.D. in Electrical and Computer Engineering

Aug 2017 - May 2023

University of Southern California

GPA: 3.93

Advisor: Antonio Ortega

M.S. in Computer Science

Aug 2017 - May 2022

University of Southern California

GPA: 4.0

M.S. in Electrical Engineering (Computer Vision, Machine Learning)

Aug 2012 - Dec 2013

University of Southern California

GPA: 3.86

B.Tech. in Electronics and Communication

July 2008 - June 2012

National Institute of Technology, Tiruchirappalli

GPA: 9.12

Work Experience

Staff Research Scientist

Oct 2024 - Present

Salesforce

Working on foundational research on (multi) agentic systems and LLM training for improved reasoning and alignment. Research on voice AI and agentic design.

Member of Technical Staff

June 2023 - October 2024

Tenyx (Acquired by Salesforce)

Part of the founding team building Voice AI for customer support. Research on continual learning, TenyxChat models, and geometric characterization of LLMs.

Research Intern

Sep 2022 - Dec 2022

Google

Sunnyvale, CA

Worked on understanding the impact of input data in training graph models and scalable sampling approaches. 3x increased recall in abuse detection.

Software Engineer 2

Mar 2014 - Oct 2016

KLA Tencor

Milpitas, CA

Designed and developed tools to classify and visualize defect modulations for Process Window Qualification in wafer fabrication.

Publications

22 publications. View all →

Echoing: Identity Failures when LLM Agents Talk to Each Other

S Shekkizhar, R Cosentino, A Earle, S Savarese, arXiv Preprints, 2025

Convergence dynamics of Agent-to-Agent Interactions with Misaligned objectives

R Cosentino, S Shekkizhar, A Earle, arXiv Preprints, 2025

AGI Is Coming... Right After AI Learns to Play Wordle

S Shekkizhar, R Cosentino, arXiv Preprints, 2025

Out-of-Distribution Detection through Soft Clustering with Non-Negative Kernel Regression

A. Gulati, X. Dong, C. Hurtado, S. Shekkizhar, S. Swayamdipta, A. Ortega, Findings of the Association for Computational Linguistics: EMNLP, 2024

Reasoning in Large Language Models: A Geometric Perspective

R Cosentino, S Shekkizhar, arXiv Preprints, 2024

Patents

Knowledge base for voice large language model applications

Provisional

US63752613 • Filed: January 2025

Gradient-free optimization of large language models

Provisional

US63752618 • Filed: January 2025

Machine learning model compression

Provisional

US18905761 • Filed: October 2024

Training a target activation sparsity in a neural network

Pending

US18802235 • Filed: August 2024

Domain aware large language model governance

Pending

US18745562 • Filed: June 2024

Fine-tuning machine learning models while retraining accumulated knowledge

Pending

US18496698 • Filed: October 2023

Data sampling using Locality Sensitive Hashing for large scale graph learning

Granted

US63517869 • Filed: August 2023

Optimizing training sets used for setting up inspection-related algorithms

Granted

US10267748 • Filed: April 2019

Awards & Honors

  • •IEEE Rising Star in Signal Processing - ICASSP 2023
  • •IEEE Best Student Paper Award - ICIP 2020
  • •Ming-Hsieh Ph.D. Scholar Finalist 2022-23

Academic Activities

  • •Reviewer: IEEE Journals (JSAIT, TSIPN, SPL, TNNLS)
  • •Reviewer: Conferences (ICASSP, ICLR, NeurIPS, LoG, ICML)
  • •Mentor: Viterbi Graduate Mentorship Program, Fall 2021
  • •VGSA Senator: Fall 2017, Spring 2020