๐Ÿ‘‹ Hi, I'm Raghvender

Iโ€™m a Ph.D. graduate in Computational Materials Science and currently serve as an AI Research Engineer at PreDeeption, a deep-tech startup pioneering Generative AI solutions for battery life-cycle prediction.

My work bridges physics and machine learning โ€” leveraging techniques like Variational Autoencoders (VAEs), diffusion models, and autoregressive transformers to model and forecast battery degradation across life cycles.

Iโ€™m driven by the challenge of turning complex scientific domains into interpretable, scalable, and elegant ML systems โ€” always with an eye toward real-world impact and sustainability.

๐Ÿงฌ Generative Modeling Toolkit

I explore and design advanced generative models to tackle challenges in battery life-cycle prediction and beyond.

Latent Variable Models
(VAE, VQ-VAE-2 and cVAE models)
Diffusion-Based Models
(DDPM-style denoising models)
AutoRegressive Models
(LSTM & Transformers over latent trajectories)

๐Ÿง  Technical Skills

Languages & Web APIs

Python
Python
C++
C++
FastAPI
FastAPI
Flask
Flask

Frameworks & Libraries

PyTorch
PyTorch
TensorFlow
TensorFlow
Scikit-Learn
Scikit-Learn
OpenCV
OpenCV
Pandas
Pandas

DevOps & Platforms

Docker
Docker
AWS
AWS
Git
Git
Linux
Linux

๐Ÿงช Work Experience

AI Research Engineer

PreDeeption (Startup) 2024 โ€“ Present

Developing deep generative models (VAE, VQ-VAE, diffusion, AR) for battery lifecycle modeling.

Research Engineer

IRCER UMR 7315 CNRS, France 2023

Physics-informed ML for computational materials modeling.

Junior Research Fellow (JRF)

S.N. Bose National Centre for Basic Sciences 2018 โ€“ 2019

DFT and atomistic simulation for materials research.

๐ŸŽ“ Education

Ph.D. in Computational Materials Science

IRCER, University of Limoges 2019 โ€“ 2023

Masters in Physics

S.N. Bose Centre / University of Calcutta 2016 โ€“ 2018

Bachelors in Physics

Sharda University 2009 โ€“ 2012

๐Ÿ“ก Stay Connected

I explore and share insights on Generative AI, ML for battery tech, and more. Follow along for clean research, modular code, and impactful ideas.