Raxit Goswami
VP Research and Development
RAAPID Inc.
About
Biography
Raxit Goswami is a research leader at the forefront of healthcare innovation, specializing in the development of Neurosymbolic AI to create safer, more interpretable medical systems. As the VP of Research at RAAPID, he leads the advancement of automated risk adjustment solutions and clinical NLP, bridging the gap between large-scale medical data and actionable insights.
By combining deep learning with symbolic reasoning, Raxit is redefining how AI supports clinical documentation and value-based care. His work focuses on building explainable systems that healthcare professionals can trust, validate, and confidently integrate into clinical practice.
Under his leadership, the RAAPID Labs research team has published extensively, filed multiple patents, and developed proprietary Neurosymbolic AI architectures that power the company’s risk adjustment and clinical coding platform.
Publications
Selected Publications
Researchteam_hcn at SemEval-2023 Task 6: A Knowledge Enhanced Transformers Based Legal NLP System
- July 2023
CRF-based Clinical Named Entity Recognition Using Clinical NLP
ezDI: A Supervised NLP System for Clinical Narrative Analysis
ezDI: A Hybrid CRF and SVM Based Model for Detecting and Encoding Disorder Mentions in Clinical Notes
Patents
Patents & Intellectual Property
NLP-Powered Language Modeling Techniques for the Clinical Domain
Patent No. 19/208,594 · Filed May 15, 2025
The present invention relates to NLP-powered language modeling techniques tailored for the clinical domain. The system leverages advanced NLP architectures to analyze, interpret, and generate clinically relevant text from structured and unstructured healthcare data. The disclosed techniques improve contextual understanding, domain adaptation, and semantic accuracy within medical narratives, clinical documentation, and healthcare workflows.
Amitava Das, Dhanachandra Singh, Raxitkumar Goswami, Pinal Patel, Amit Sheth
Geometric Reprojection Instruction Tuning for Language Model Adaptation
Non-provisional due December 17, 2026
The present invention relates to advanced geometric reprojection techniques for parameter-efficient instruction tuning of large language models. The disclosed system introduces curvature-aware optimization and subspace projection methods to adapt pre-trained models to downstream tasks while minimizing computational overhead.
Amitava Das, Raxitkumar Goswami
Neurosymbolic AI System for Automated Medical Coding and Risk Adjustment
Patent No. 63/956,664 · Non-provisional due January 8, 2027
The present invention relates to curvature-aware parameter updates and dynamic subspace projection mechanisms to enhance fine-tuning performance while maintaining computational efficiency. The method selectively updates model parameters in optimized subspaces derived from the loss landscape geometry.
Amitava Das, Raxitkumar Goswami
Method for Visual Paraphrase Attack Safe and Distortion Free Image Watermarking for AI-Generated Images
Patent No. 63/921,281 · Non-provisional due November 20, 2026
The present invention relates to a method for visual paraphrase attack-safe and distortion-free image watermarking for AI-generated images. The system embeds robust and imperceptible watermarks into synthetic images produced by generative models, ensuring resistance against adversarial manipulation.
Amitava Das, Raxitkumar Goswami
Research Interests
Areas of Focus
Neuro-Symbolic AI for Healthcare
Combining neural networks with symbolic reasoning to build interpretable, explainable AI systems for clinical applications and risk adjustment.
Automated Risk Adjustment & Clinical NLP
Developing automated systems for clinical coding, HCC capture, and RADV audit defense using advanced natural language processing.
Knowledge-Infused Machine Learning
Integrating structured domain knowledge and medical ontologies into ML pipelines to improve accuracy, reliability, and regulatory compliance.
Other Researchers
Pinal Patel
Director of R&D
Disha Davey
Dir. Clinical Informatics
Dhanachandra N.
Research Team Lead
Dr. Amit Sheth
Advisor, IAIRO
Dr. Amitava Das
Advisor, BITS Pilani
Anurag Deo
Assoc. Research
Engineer