Suggestions
Mayank Malik
Chief Data Officer and Blockchain Lead at SLAC National Accelerator Laboratory. Not active on LinkedIn.
Mayank Malik is an accomplished software engineer currently working at Google since July 2022. He has a strong background in software engineering, machine learning (ML), and deep learning, with a notable history of developing complex systems and algorithms in various roles.
Professional Experience
-
Google: Software Engineer (July 2022 - Present)
-
Amazon: Software Development Engineer (December 2020 - July 2022)
- Developed a Delivery Route System for Amazon Go and Amazon Kitchen stores, optimizing delivery truck schedules using backend technologies such as Java and Python.
-
Deloitte: Consultant in Software and AI/ML Practice (February 2020 - December 2020)
- Worked on machine learning models for pharmacy retail, enhancing operational efficiency.
-
American Express Global Business Travel: Assistant Manager (April 2014 - August 2019)
- Led initiatives to develop machine learning models that significantly improved client retention and revenue.
-
Accenture: Software Engineer Analyst (2010 - 2013)
Education
-
Northwestern University: Master’s in Artificial Intelligence & Machine Learning (2018 - 2019)
-
Thapar Institute of Engineering & Technology: Bachelor of Engineering in Electronics and Communication (2006 - 2010)
Malik has also contributed to various projects involving AI and machine learning, including developing algorithms for predicting Parkinson's Disease from MRI data and creating a gesture-controlled game using computer vision techniques.1
Highlights
The discrepancy between LLMs' ability to excel in Math Olympiads and struggle with basic arithmetic questions like 'which number is bigger, 9.11 or 9.9?' is puzzling. One explanation is in training data distribution. While AlphaProof and AlphaGeometry-2 were trained on formal proofs and domain-specific data, GPT-4o had more diverse training data possibly including data from Github's public repositories. In software versioning, v9.11 is indeed larger than v9.9, which may have skewed the training data distribution, and resulted in GPT-4o "thinking" 9.11>9.9.
https://lnkd.in/gdtTEbkh https://lnkd.in/gz-gFsmE
Congratulations to Boyi Li, Chinmay Gadgil, Dharini Krishna, Jack Ryan, Joe Chang, Rahul Raja, Shiva Amrit Avasarala, Tianming (Robert) Wang, Wenjie Y., Yizhou Wei, YueBin Yang for a successful poster presentation at CMU INI Practicum. Thank you for everything you've done and continue to do to further our mission of 100% clean energy for all.
Special thanks to CMU faculty and staff, and SLAC mentors for their tremendous support. Aleecia McDonald, Bennet Meyers, davide innocenti, Dena Haritos Tsamitis, Gregory Kesden, Gustavo Cezar, Jayshree Ranka, Jennifer Feyrer, Osman Yagan, Patrick Tague, Pei Zhang
