About Research

Shailja Thakur

I am Research Scientist at IBM Research where my focus is on LLMs for Code, Code Explanation, Reasoning & Alignment. My primary research focuses on applying AI to develop efficient, reliable, and transparent systems. Alongside, I work on AI for social Impact. I have a PhD from University of Waterloo advised by Sebastian Fiscmeister and a PostDoc from New York University, working with Siddharth Garg and Ramesh Karri.

profile photo

Linkedin Google Scholar Twitter Github HuggingFace

Recent Updates & Highlights
  • Serving on Program Committee for AAAI 2025, Artificial Intelligence for Social Impact Track Program Committee
  • Serving on Program Committee for WiML 2024
  • Serving on Program Committee for EMNLP 2024, Workshop for Multilingual Representations (MRL)
  • Our paper VeriGen got accepted in ACM TODAES
  • Our two papers on (1) bug fixing and, (2) assertion generation using LLM got accepted in IEEE TIFS
  • Invited Talk, Speaker Series: AI and Security CoP, Talk title: Automating Verilog RTL Code Generation with Large Language Models Intel, New York 2023
  • LLMs for hardware design, Tutorial Title: Intersection of HW, Security & Large Language Models @MLCAD 2023
  • Invited Reviewer for WiCV at ICCV 2023
  • Invited Reviewer for AAAI 2023
  • Nominated for Best paper award Benchmarking LLM for Verilog RTL generation DATE 2023
  • Travel grant for DATE 2023
  • Granted Post-Doctoral Fellowship in the NSF National AI Institute for Edge Computing Leveraging Next Generation Networks (Athena) 2022
Invited Talks, Research Seminars & Workshop Sessions
  • IIIT Hyderabad, Talk Title: Large Language MOdel for Hardware Design Automation, September 19, 2024
  • PALS Industrial Assisted Lecture Series, Talk Title: Building Successful LLM Applications, Power of High-Quality Data, August 27, 2024
  • MSRIT Invited Talk, Topic: Large Language Models for Code Generation, Reasoning and Preference Alignment
  • Invited Seminar at CDS@IISc, CSE Dept@IIT Kanpur, IISER Pune, IIIT Delhi, IIT Mandi on Towards Secure, Interpretable, and Scalable Machine Learning Applications in Cyber-Physical Systems, Feb-September 2024
  • Invited Talk, Speaker Series: AI and Security CoP, Talk title: Automating Verilog RTL Code Generation with Large Language Models Intel, New York 2023
  • LLMs for hardware design, Tutorial Title: Intersection of HW, Security & Large Language Models @MLCAD 2023
  • CEDAR, INRIA and Ecole Polytechnique on Title: Security and interpretability in automotive systems, January 2022
  • The School of Computing and Augmented Intelligence, Arizona State University Title: Security and interpretability in automotive systems, October 2021
  • the Cyber-Physical Systems Lab, Duke Universiy on Title: Security and interpretability in automotive systems, November 2021

Contact & Presence

Email   Discord   Twitter   Github   HuggingFace