Suyash Maniyar I am a Master’s student in Computer Science at University of Massachusetts Amherst. In my first semester, I am doing coursework in Advanced Natural Language Processing, Software Engineering and conducting research under Prof. Hong Yu at the BioNLP Lab. My ongoing work focuses on ontology-driven self-play for Biomedical LLMs, where I train large language models on biomedical corpora (PubMed, MIMIC) with ontology-grounded QA loops. smaniyar@umass.edu  /  Download Resume (PDF)  /  LinkedIn  /  GitHub

📄 Publications

• AI-Generated Lecture Slides for Improving Slide Element Detection and Retrieval, ICDAR 2025.
  
  
  [Project Website]

💼 Experience

Advanced Data Science Associate — ZS Associates (Nov 2024 – Jul 2025)

• Developed and deployed a Switch Prediction Model using ML/DL on real-world data (projected savings of $500K).
• Performed Patient Archetyping and Patient Journey modeling using SQL, GMM, XGBoost, Autoencoders.

Data Scientist — Decimal Point Analytics (Jul 2023 – Oct 2024)

• Developed RAG systems for enterprise documents, reducing analyst query time by 93%.
• Integrated knowledge graphs into RAG chatbot (+9% accuracy, –75% latency).
• Built predictive maintenance models for industrial machinery.
• Trained diffusion models for generating fabric designs from text prompts.

Artificial Intelligence Researcher — RaapidAI X IIT Jodhpur (Aug 2022 – May 2023)

• Worked on table detection and extraction from medical documents using CascadeTabNet, DETR, DIT (mAP 76%).

Data Science Intern — Decimal Point Analytics (May 2022 – Jul 2022)

• Built accident detection and car damage segmentation pipelines using ViT, ConvLSTM, YOLO, DeepSORT (F1 = 87%).