Shubhro Dev

Bengaluru, India

shubhro2004@gmail.com

+91 8272979689

Hi, I am Shubhro Dev, a final-year Computer Science and Engineering undergraduate at the Rajiv Gandhi Institute of Petroleum Technology (RGIPT), an Institute of National Importance built along the lines of IITs. I work at the intersection of large language models, multimodal deep learning, federated learning and time-series analysis, and I care most about building models that are not just accurate but also verifiable and privacy preserving.

I am currently a Technical Research Intern with the Global DAI R (SDE-IN) team at Siemens, researching the validation and verification of outputs generated by LLMs. Before this I spent time as a research assistant and intern at labs including VANET Lab (IIT Jodhpur), VCA Lab (IIT BHU Varanasi) and CMATER Lab (Jadavpur University), and as a data science and machine learning intern at Core-AI Solutions and Adani Power. Along the way I have worked on predictive maintenance, medical imaging, and multimodal knowledge retrieval.

My work has led to publications in medical imaging and privacy-preserving predictive maintenance, and I placed at Global Rank 10 in the Meta Hacker Cup 2025 (AI Track). When I am away from research, you will usually find me playing chess, building software tools, or gaming.

Feel free to explore my publications, projects and CV, or reach out over any of the links below.

news

Jan 05, 2026	Started as a Technical Research Intern with the Global DAI R (SDE-IN) team at Siemens, working on the validation and verification of LLM outputs.
Nov 29, 2025	Global Rank 10 at Meta Hacker Cup 2025 (AI Track)
Jan 17, 2025	Our paper on lung cancer identification from CT scans using a soft-attention deep transfer learning model was accepted at IEEE ISACC 2025.

selected publications

ISACC
Lung Cancer Identification from CT Scans using a Soft-attention enabled Deep Transfer Learning Model

Shubhro Dev, P. S. Roy, N. Chakraborty, and 1 more author

In 2025 3rd International Conference on Intelligent Systems, Advanced Computing and Communication (ISACC), 2025

Abs DOI Bib PDF

A soft-attention enabled deep transfer learning model for identifying lung cancer from CT scans and histopathological images, built on transfer learning and attention mechanisms for robust multi-source classification.
@inproceedings{dev2025lungcancer, title = {Lung Cancer Identification from CT Scans using a Soft-attention enabled Deep Transfer Learning Model}, author = {Dev, Shubhro and Roy, P. S. and Chakraborty, N. and Sarkar, Ram}, booktitle = {2025 3rd International Conference on Intelligent Systems, Advanced Computing and Communication (ISACC)}, pages = {254--259}, year = {2025}, doi = {10.1109/ISACC65211.2025.10969319}, }
IEEE T-ITS

A Differentially Private Hybrid Transformer-Mamba Architecture for Industrial Predictive Maintenance on Multivariate Time Series Data

Shubhro Dev, A. Ghosh, and Debasis Das

IEEE Transactions on Intelligent Transportation Systems. In submission , 2026

Abs

A privacy-preserving hybrid Transformer-Mamba architecture for remaining useful life prediction and industrial predictive maintenance on multivariate time-series data, combining a custom time-series encoder with differential privacy.