CV
Education
- Ph.D. in Computer Science, University of Minnesota, Twin Cities, 2018–2023
- Advisor: Prof. Vipin Kumar
- Dissertation: “Entity-Aware Knowledge-Guided Machine Learning for Scientific Knowledge Discovery”
- B.Tech in Electronics & Electrical Engineering, Indian Institute of Technology (IIT) Guwahati, 2012–2016
Work Experience
- Applied Scientist — Amazon AWS GenAI Innovation Center, Santa Clara (Jul 2024 – Present)
- CompAgent (CVPR’26): Agentic framework for visual compliance verification using multi-modal LLMs
- TelcoAI (AACL-IJCNLP’25): Agentic multi-modal RAG system for 3GPP technical specification search
- LLM Evaluation: Probabilistic frameworks for evaluating LLM self-consistency and logical reasoning (NeurIPS’25 Workshop, ICLR’26 Workshop)
- Applied Scientist — Amazon Music, San Francisco (Aug 2023 – Jul 2024)
- Causal inference frameworks for measuring treatment effects of recommendation systems at scale
- Foundational representation model for Amazon Music enabling personalization, content understanding, and discovery
- Graduate Research Assistant — University of Minnesota (Sep 2018 – Aug 2023)
- Spatio-temporal foundation models using UNet architectures (LSTM + Attention) for satellite image time-series segmentation
- Inverse modeling of dynamical systems using knowledge-guided self-supervised learning (KDD’22, SDM’24, ICDM’25)
- Large-scale remote sensing: First high-resolution map of Arctic lakes, palm oil plantation maps (Indonesia), cashew plantation maps (Africa)
- Applied Scientist Intern — Amazon Music, San Francisco (Jun 2022 – Sep 2022)
- Cross-domain recommender leveraging customer behavior from video/books to recommend Podcast/Music content
- Applied Scientist Intern — Amazon Music, San Francisco (Jun 2021 – Sep 2021)
- Cold-start podcast recommender using propensity models and cross-domain behavioral signals
- Research Engineer — Samsung R&D Institute, Delhi, India (Jul 2016 – Jun 2018)
- Time-series fault prediction on IoT sensor data for Samsung digital appliances
- Smart CRM engine using NLP for defect extraction and diagnosis recommendation
- Research Intern — Reserve Bank of India R&D Institute, Hyderabad (May 2015 – Jul 2015)
- Quantum-inspired differential evolution for feature subset selection; deep learning for sentiment analysis
Research Interests
- Core: Foundation Models, Sequence Modeling, Computer Vision, Multi-Modal Modeling, Representation Learning
- Methods: Self-supervised Learning, Few-shot & Zero-shot Learning, Meta Learning, Variational Inference
- Applications: Uncertainty Quantification, Causal Inference, Scientific AI, Earth Observation
Honors & Awards
- IEEE ICDM Best Paper Award (2021, 2023)
- ACM SIGSPATIAL Best Paper Award (2022)
- SIAM SDM Best Paper Award (2022)
- DeepSpatial @ KDD Best Paper Award (2021)
- 3M Fellowship (2018)
- MnDrive UMII Fellowship (2020)
- HydroML Travel Award (2022)
- KDD Student Travel Award (2019, 2020)
- All India Rank 6 — International Masters Mathematics Olympiad (2011)
- Top 0.5%ile — IIT Joint Entrance Exam (IIT-JEE) (2012)
Patents
- Liu, C., Ghosh, R., Asokkumar, A., Rele, G., Yadav, A., Ravipati, V. Visual Compliance Verification using Agentic Multi-Modal Large Language Models. U.S. Patent App P90445-US01.
- Ghosh, R., Ravirathinam, P., Mulla, D. and Kumar, V. (2023). Inverse Modeling based approach for Land Cover Mapping. U.S. Patent App 63/605,220.
Publications
Software & Datasets
- ReaLSAT — A global dataset of reservoir and lake surface area variations. Nature Scientific Data, 9(1), 1–12, 2022.
Professional Service
- Program Committee: NeurIPS (2022–2024), ICML (2022–2024), ICLR (2021–2022), AAAI (2022–2023), AAMAS (2021), IJCAI (2020)
- Journal Reviewer: PNAS, Nature Machine Learning, Nature Communications, Water Resources Research
- Mentorship: Supervised 5 undergraduate and 3 graduate students on research projects
