Akash Srivastava
Scholar

Akash Srivastava

Google Scholar ID: 2h6SZeEAAAAJ
IBM, MIT, Red Hat AI, IBM Research, University of Edinburgh
Generative AILarge Language ModelsDensity Ratio EstimationSelf-Supervised Learning
Citations & Impact
All-time
Citations
2,989
 
H-index
20
 
i10-index
28
 
Publications
20
 
Co-authors
128
list available
Resume (English only)
Research Experience
  • PI at MIT-IBM AI Lab (2023–present)
  • Leads multiple projects including: generative modeling for constrained mechanical systems, synthetic data in decision-making, generative active learning for atomistic simulators, and rethinking vehicle routing with modern ML
  • Collaborates with MIT faculty including Faez Ahmed, Dean Eckles, Rafael Gomez-Bombarelli, and Matthias Winkenbach
  • PI for Climate Change AI project (2022): optimizing multi-modal urban delivery for greener last-mile logistics
  • Co-PI on DARPA’s Machine Common Sense project (2019)
Background
  • Principal Investigator (PI) at the MIT-IBM AI Research Lab
  • Chief Architect of Large Language Model Alignment at IBM Research
  • Oversees tuning and alignment of IBM’s foundation models
  • Leads synthetic data generation efforts
  • Lab located on MIT campus at 314 Main St (new MIT Museum building)