Eashan A.

Saff Researcher, AI Model @ Lenovo

About

My work is inclined in the area of efficient perception in multimodal LLMs. My goal is to make the conceptual advances necessary for agentic systems to be fast, reliable and aligned with human values.Homepage: https://eashanadhikarla.github.ioGoogle Scholar: https://scholar.google.com/citations?user=k3BMw_QAAAAJ&hl=en&oi=aoSpecialities: Local LLMs, Agentic Systems, Multimodal Perception, On-device Modeling, Efficient Foundation Models

Country

United States

City

Chapel Hill

Industry

Research

Skill

MLLM, Agentic AI Development, Local LLMs, On-device SLM, Large Language Models (LLM), Generative AI, PyTorch, Python (Programming Language), C++, Python, Machine Learning, Data Mining, Programming, Software Development, Research, Algorithms, Project Management, Network Security, Artificial Intelligence (AI), Computer Vision

Experience

Saff Researcher, AI Model

Lenovo

2026-5 - Present · 5 mos

Morrisville, NC

Office of CTO

Staff Researcher Intern

Lenovo

2022-6 - 2024-11 · 2 yrs 6 mos

Chicago, Illinois, United States

This role involves both part-time and full-time collaboration with multiple teams during my PhD.

PhD Summer Intern

Berkeley Lab

2021-5 - 2021-8 · 4 mos

Berkeley, CA

Data Science Intern

Berkeley Lab

2019-5 - 2019-8 · 4 mos

Berkeley, California Area

AI/ML Intern

Persistent Systems

2016-7 - 2016-9 · 3 mos

Pune/Pimpri-Chinchwad Area

Education

Lehigh University

Computer Science

2020-8 - 2026-3 · 5 yrs 8 mos

Advisor: Dr. Brian D. Davison Focus: Efficient Generative Models, On-device Image Generation, Small Language Models Graduate Teaching Assistant Aug 2018 - May 2022 · 3 yrs 10 mos Bethlehem, Pennsylvania • CSE160: Introduction to Data Science - (Spring 2022, Fall 2021, Spring 2019) • CSE017: Programming and Data Structures - (Fall 2020, Fall 2019) • CSE001: Breadth of Computing - (Fall 2018) Graduate Research Assistant Jan 2021 - May 2021 · 5 mos Peer Mentor (at REU Lehigh) May 2020 - Aug 2020 · 4 mos Funded by National Science Foundation (CNS-1757787) Graduate Research Assistant Aug 2019 - May 2020 · 10 mos Bethlehem, Pennsylvania, United States Advisor - Prof. Brian D. Davison • Designing a robust auto-encoder and GAN for detecting adversarial images. • Developed a close proximity approximation model which is also known as on manifold adversarial detectors. • Enhanced the vanilla auto-encoder to a deep architecture with enforced learning from memory elements.