May 2024 – Jul 2024 · 2 mos · Gurugram, Haryana, India · On-site

• Worked on alpha research in US, European and Japanese Stock Market (Equity Space)
• Developed a robust and high performance trading strategy based on a new class of alphas in Japanese Stock Market
• Developed novel News Sentiment based data variables in Equity Space
• Worked on novel feature engineering for a trading strategy

Aug 2023 – Jun 2025 · 1 yr 10 mos · Kharagpur, West Bengal, India

• Working on Conversational AI Systems

May 2023 – Jul 2023 · 2 mos · Edmonton, Alberta, Canada · On-site

• Mentor: Prof. Dana Cobzas
• Worked on the segmentation of hippocampi for people suffering from diseases like Temporal Lobe Epilepsy, Parkinson's Disease, etc.
• Developed a model for encoding shape in the latent space of our NestedResUNet segmentation model to perform efficient segmentation.
• Proposed model has 1.4M parameters as compared to 10.4M in UNet++ yet achieves a dice score of 86.84 (4% increment over baseline).

Dec 2022 – Feb 2023 · 2 mos · Kolkata, West Bengal, India

• Mentor: Soma Bandyopadhyay (TCS Innovation Lab Kolkata)
• Worked on Synthetic Time Series Data Generation for both univariate & multivariate time series data in domains like finance, IoT, etc.
• Designed Class-Conditioned Time Variational Autoencoder (cTimeVAE) for backtracking class labels during synthetic data generation.
• Improved the performance on 4 benchmarking real-world datasets from the UCI/UCR Repository by reducing the discriminative classifier scores between the original and generated dataset by 0.04-0.12 over the original Time Variational Autoencoder model.

May 2022 – Jun 2022 · 1 mo · Kolkata, West Bengal, India

• Mentor: Prof. Utpal Garain (ISI Kolkata) & Dr. Subhasish Sarkar (Sagore Dutta Hospital)
• Worked on Cancer Research through Image Processing tools and frameworks in collaboration with Sagore Dutta Medical College and Hospital.

Apr 2022 – May 2023 · 1 yr 1 mo

Oct 2021 – Apr 2022 · 6 mos

Apr 2022 – May 2023 · 1 yr 1 mo · Australia

• Working on Explainable Uncertainty Quantification Methods

Dec 2021 – Jul 2022 · 7 mos · Patna, Bihar, India

• Mentor: Prof. Asif Ekbal & Kshitij Mishra (PhD)
• Proposed a novel persona-consistent polite persuasive dialogue system by designing an efficient PPO loss-based reward function.
• Improved performance over state-of-the-art models ARDM and PersRFI by reducing perplexity value from 12.45 & 12.38 to 5.52.
• Enhanced the average dialogue length from 15.03 & 19.36 to 23.83 and the persuasive strategy score from 49.2% & 51.2% to 57.3%.

Aug 2021 – Sep 2021 · 1 mo

• Worked on Front-end application development using React Native and related technology stack and Integration of Front-end and Back-end.