Shreyas Vinaya Sathyanarayana

I’m an AI Research Scientist at MStack AI, where I design Chemistry-first AI models for molecule synthesis and chemical supply chains. I completed my Bachelor of Engineering in Computer Science and Masters of Science in Chemistry from BITS Pilani, Goa Campus.

Previously, I was a Research Fellow at Deep Forest Sciences, where I led the development of DeepRetro — a novel AI framework for retrosynthesis that achieved state-of-the-art results on multi-step synthesis benchmarks. This work was awarded the $100,000 Standard Industries Chemical Innovation Challenge. I completed my master’s thesis at Yale University with Prof. Smita Krishnaswamy, working on geometric deep learning for molecule generation.

I’m an active contributor to DeepChem, the open-source drug discovery toolkit, and serve as a Google Summer of Code mentor. At BITS Pilani, I worked with Prof. Ashwin Srinivasan at APPCAIR on AI for drug discovery.

My research interests span AI for Chemistry, Drug Discovery, Graph Neural Networks, Large Language Models, and Computational Biology.

news

Jan 1, 2026	Our paper “DeepRetro: Retrosynthetic Pathway Discovery using Iterative LLM Reasoning” has been accepted at Nature Scientific Reports! 🎉
Jan 1, 2026	“ChemBERTa-3: An Open Source Training Framework for Chemical Foundation Models” has been accepted at Royal Society of Chemistry’s Digital Discovery!
Aug 1, 2025	Started as an AI Research Scientist at MStack AI, working on Chemistry-first AI models for molecule synthesis and chemical supply chains! 🚀
Jun 1, 2025	Our paper “DeepChem-Variant: A Modular Open Source Framework for Genomic Variant Calling” has been accepted at the Championing Open-source DEvelopment in ML Workshop @ ICML 2025!
May 1, 2025	Serving as a Google Summer of Code Mentor for DeepChem for the second year! Mentoring projects on polymer generative pipelines and model migration.
Dec 1, 2024	🏆 Won the Standard Industries Chemical Innovation Challenge with a prize of $100,000 for our work on DeepRetro!
Oct 11, 2024	Our work “Open-Source Molecular Processing Pipeline for Generating Molecules” has been accepted at Machine Learning and the Physical Sciences workshop happening at NeurIPS 2024
Jun 21, 2024	I have joined Krishnaswamy Lab at Yale University as a Visiting Researcher Intern. I will be working on Molecular Generation, Graph signal processing and Deep Learning.
May 27, 2024	I have joined Sprinklr as a Product Engineering Intern in 2024. I will be working with the Voice team in the Machine Learning department. I am really excited to work with them!
Apr 13, 2024	Won third prize in ACM Goa Chapter’s event PitchIt! organized at BITS Goa for my research presentation. You can find the slides here
Jan 20, 2024	I will be a TA for the course “BITS F464 - Machine Learning” under Prof. Aditya Challa and Prof. Ashwin Srinivasan this semester. (You can check out our labs here)
Jan 1, 2024	I have joined Kreiman Lab at Harvard Medical School as a Research Intern. I will be working with Prof. Gabriel Kreiman on Computational Neuroscience, Deep Learning and Natural Language Processing.
Dec 10, 2023	My paper “Predicting ATP binding sites in protein sequences using Deep Learning and Natural Language Processing” got accepted at AI2SE workshop at AAAI 24
Aug 15, 2023	Started working at APPCAIR as a student researcher!
Aug 15, 2023	I will be an instructor for the CTE course “Intro to ML/DL” with Karan and Tejas, this semester.
Jul 1, 2023	Only one from our batch to get inducted into SAiDL!
Jun 1, 2023	I have joined Deep Forest Sciences as an Intern. I will be working on Deepchem.
Jun 1, 2022	I have joined Skan.ai as a Data Science Intern.
Jun 1, 2022	I have joined CEERI Chennai as an Intern. I will be working on rPPG on resource constrained devices.

latest posts

Dec 4, 2023	Principles of Programming Languages notes

selected publications

Nat Sci Rep
DeepRetro: Retrosynthetic Pathway Discovery using Iterative LLM Reasoning

S. V. Sathyanarayana, S. D. Hiremath, R. Shah, and 4 more authors

Nature Scientific Reports, 2026

Abs Bib

DeepRetro is a novel, open-source hybrid AI framework that solves complex retrosynthesis problems by integrating the generative power of LLMs with the precision of traditional template-based engines. Achieved state-of-the-art performance with 96.3% success rate on the USPTO-190 multi-step benchmark.
@article{deepretro2026, title = {DeepRetro: Retrosynthetic Pathway Discovery using Iterative LLM Reasoning}, author = {Sathyanarayana, S. V. and Hiremath, S. D. and Shah, R. and Panda, R. and Jana, R. and Singh, R. and Ramsundar, B.}, journal = {Nature Scientific Reports}, year = {2026}, category = {journal} }

RSC DD

ChemBERTa-3: An Open Source Training Framework for Chemical Foundation Models

R. Singh, A. A. Barsainyan, R. Irfan, and 15 more authors

Royal Society of Chemistry’s Digital Discovery, 2026

Bib

@article{chemberta3_2026,
  title = {ChemBERTa-3: An Open Source Training Framework for Chemical Foundation Models},
  author = {Singh, R. and Barsainyan, A. A. and Irfan, R. and Amorin, C. J. and He, S. and Davis, T. and Thiagarajan, A. and Sankaran, S. and Chithrananda, S. and Ahmad, W. and Jones, D. and McLoughlin, K. and Kim, H. and Bhutani, A. and Sathyanarayana, S. V. and Viswanathan, V. and Allen, J. E. and Ramsundar, B.},
  journal = {Royal Society of Chemistry's Digital Discovery},
  year = {2026},
  category = {journal}
}

IEEE I2CT

Speech to Equation Conversion using a PoE Tagger

Peeta Basa Pati, and V Shreyas

In 2022 IEEE 7th International conference for Convergence in Technology (I2CT), 2022

Bib

@inproceedings{PoETagger,
  author = {Basa Pati, Peeta and Shreyas, V},
  booktitle = {2022 IEEE 7th International conference for Convergence in Technology (I2CT)},
  title = {Speech to Equation Conversion using a PoE Tagger},
  year = {2022},
  pages = {1-4},
  doi = {10.1109/I2CT54291.2022.9824252},
  category = {conference},
  url = {https://ieeexplore.ieee.org/document/9824252}
}

NeurIPS WS

Open Source Molecular Processing Pipeline for Generating Molecules

V Shreyas, Jose Siguenza, Karan Bania, and 1 more author

In Machine Learning and the Physical Sciences Workshop at NeurIPS 2024, 2024

Also presented at MoML 2024 and Baylearn 2024

Bib

@inproceedings{openMolGen2024,
  title = {Open Source Molecular Processing Pipeline for Generating Molecules},
  author = {Shreyas, V and Siguenza, Jose and Bania, Karan and Ramsundar, Bharath},
  booktitle = {Machine Learning and the Physical Sciences Workshop at NeurIPS 2024},
  year = {2024},
  eprint = {2408.06261},
  archiveprefix = {arXiv},
  primaryclass = {cs.LG},
  doi = {10.48550/arXiv.2408.06261},
  category = {workshop},
  url = {https://arxiv.org/abs/2408.06261},
  note = {Also presented at MoML 2024 and Baylearn 2024}
}

AAAI WS

Predicting ATP binding sites in protein sequences using Deep Learning and Natural Language Processing

V Shreyas, and Swati Agarwal

In 3rd Annual AAAI Workshop on AI to Accelerate Science and Engineering (AI2ASE), 2024

Bib

@inproceedings{ATP-binding2024,
  title = {Predicting ATP binding sites in protein sequences using Deep Learning and Natural Language Processing},
  author = {Shreyas, V and Agarwal, Swati},
  booktitle = {3rd Annual AAAI Workshop on AI to Accelerate Science and Engineering (AI2ASE)},
  year = {2024},
  eprint = {2402.01829},
  archiveprefix = {arXiv},
  primaryclass = {q-bio.BM},
  doi = {10.48550/arXiv.2402.01829},
  category = {workshop},
  url = {https://arxiv.org/abs/2402.01829}
}