Abhinav Gupta

Abhinav Gupta

Senior ML Scientist at Sanofi | MIT PhD in Mechanical Engineering and Computation

Sanofi

MSEAS Lab

Massachusetts Institute of Technology

Biography

Working at Sanofi to apply cutting-edge computation, AI/ML/DL, and structural biology technologies to resolve challenges in real-world drug discovery.

As a part of my PhD at MIT, I developed state-of-the-art algorithms and methodologies for uncertainty quantification, Bayesian learning, deep learning, and numerical methods for predictive ocean applications. The algorithms developed in my research were problem agnostic, and could be widely applied.

My unique background in mechanical engineering, computational biology, applied mathematics, and computing position me to identify academically understood cross-disciplinary solutions and translate them to solve real-world challenges.

Interests

  • Computational Biology
  • Deep Learning
  • Bayesian Modeling and Inference
  • Uncertainty Quantification
  • Ocean Modeling
  • Optimal Sampling
  • Numerical Methods

Education

  • PhD in Mechanical Engineering and Computation, 2022

    Massachusetts Institute of Technology

  • Master's and Bachelor's in Mechanical Engineering, 2016

    Indian Institute of Technology Kanpur

Skills

Python

70%

Machine Learning

100%

Cooking

150%

Experience

 
 
 
 
 

Senior Machine Learning Scientist, Next Generation Biologics Design

Sanofi

Aug 2022 – Present Cambridge, MA
Working in an interdisciplinary team with the Sanofi Large Molecule Research Platform to apply cutting-edge computation, AI/ML/DL, and structural biology technologies to resolve challenges in real-world drug discovery and make impacts on patients’ life.
 
 
 
 
 

Graduate Teaching Assistant

Department of Mechanical Engineering, MIT

Feb 2018 – May 2018 Cambridge, MA
Teaching assistant for the course “Numerical Fluid Mechanics (2.29)”. Held recitations every week, helped prepare lecture material, graded assignments, and quizzes. Mentored approx. ~ 25 for their end-of-term projects.
 
 
 
 
 

Research Assistant

Multidisciplinary Simulation, Estimation, Assimilation Systems (MSEAS) Lab, MIT

Jan 2017 – Jul 2022 Cambridge, MA

Research Profile:

  • Advancing algorithms on the intersection of uncertainty quantification, Bayesian modeling and inference, machine learning, and data-driven scientific computing for high-dimensional and multidisciplinary problems
  • Developing a partial-differential-equations-based Bayesian machine learning framework for model discovery; has applications in learning ocean ecosystem models, sustainable fisheries management, brain tumor modeling and more
  • Developed a delay-differential-equations based deep learning framework to learn closure models for dynamical systems; applications include capturing the effects of subgrid-scale processes in coarse models, simplification of complex biochemical models, and more

Allied Contributions:

  • Developing collaboration protocols to facilitate multi-university-research project across 5 universities
  • Mentored 3 undergraduate and 3 high-school interns on research projects
 
 
 
 
 

International Visiting Student

Multidisciplinary Simulation, Estimation, Assimilation Systems (MSEAS) Lab, MIT

May 2015 – Jul 2015 Cambridge, MA
Gained skills in numerical methods, probabilistic modeling, and data-assimilation.
 
 
 
 
 

International Visiting Student (S.N. Bose Scholar)

Multidisciplinary Simulation, Estimation, Assimilation Systems (MSEAS) Lab, MIT

May 2015 – Jul 2015 Cambridge, MA
Gained skills in numerical methods, and ecosystem modeling for the ocean.
 
 
 
 
 

Summer Undergraduate Research Grant for Excellence (SURGE) Fellow

Indian Institute of Technology Kanpur

May 2015 – Jul 2015 Kanpur, India
  • Analyzed the phenomenon of vortex shedding and its control using active and passive methods
  • Gained experience in computational fluid dynamics (CFD) and numerical methods

Accomplish­ments

Wunsch Foundation Silent Hoist and Crane Award

Awarded for Outstanding Graduate Student

Best Poster Award

For the poster titled “Neural Closure Models for Dynamical Systems” at the “Annual MIT CCSE Symposium”

MathWorks Mechanical Engineering Fellowship

Awarded to 3 out of 500 graduate students for exceptional academic performance

SIAM Student Travel Award

For attending the Uncertainty Quantification 2020 conference.

MIT-Tata Center for Technology & Design Fellow

Studied interplay of technology, entrepreneurship, and policy; and deepened perspectives on severely resource-constrained communities by interviewing Indian fishermen, non-profits, and government institutions.

General Proficiency Medal | Banco Foundation Award | OP Bajaj Memorial Award

For the best academic performance in the department of Mechanical Engineering.

S.N. Bose Scholarship

Offered to select Indian studnets to experience world-class research facilities in leading U.S. institutions.

Projects

*

Neural Closure Models for Dynamical Systems

Developed a novel, versatile, and rigorous methodology to learn non-Markovian closure parameterizations for low-fidelity models using data from high-fidelity simulations.

Active rank-1 updates to POD based reduced order models

Investigation of a method for efficiently updating a reduced order model online, as new data becomes available corresponding to system changes.

Multiscale Gaussian Process Regression

Investigating weighted mix-ture of GPs method for handling multiscale features.

Physics-Inspired Machine Learning for PDEs

Implementation of Machine Learning algorithms for solving nonlinear partial differential equations, in a hybrid framework.