Applied Scientist
Building with AI, One Step at a Time
Data Scientist and MS student at IIIT Hyderabad, focusing on NLP and Vision-Language Models for document understanding and reasoning.
Data Scientist and MS student at IIIT Hyderabad, focusing on NLP and Vision-Language Models for document understanding and reasoning.
I am a Data Scientist at Perfios Software Solutions and currently pursuing an MS in Data Science at IIIT Hyderabad. My work primarily involves using NLP and Vision-Language Models to help machines better understand documents.
I enjoy working across the machine learning lifecycle, turning research ideas into practical applications. My open-source work on the Kuvera personal finance models has been downloaded over 51,000 times on Hugging Face.
Perfios Software Solutions
Experimenting with VLMs (like PaliGemma2) for financial data, building reasoning workflows, and developing algorithms for document readability.
Perfios Software Solutions
Reduced inference latency 8s→200ms via distillation; improved table detection by 27.6%; enhanced TSR with semantic row-detection.
IIT Kharagpur
Technical articles on machine learning, reasoning systems, and applied research.
Analyzing dense vs diverse sampling strategies for VLM training on 15k-sample synthetic datasets.
Read More →How I ranked 1st globally in the Reasoning Dataset Creation Challenge using synthetic data.
Read More →An overview of how training regimes evolved from classic approaches to modern reasoning models.
Read More →Lessons learned from production PyTorch development covering best practices and optimization.
Read More →Ranked 1st globally among 150+ teams (Bespoke Labs, HuggingFace, Together.ai). Trained a 7B model that achieved strong reasoning capabilities.
Instruction-tuning dataset for Indian financial context. Fine-tuned 8B/14B models with 51,000+ downloads.
An interactive tool for reading research papers, using a tiered approach to balance speed and depth when analyzing documents.
Won the Circle of Excellence award at Perfios (2026). Creator of Kuvera datasets and models with 51,000+ downloads.
AAAI 2026 Workshop (FinForge) and FinNLP @ EMNLP 2025 on behaviour-aware personal finance LLMs.
The Reasoning Course (HuggingFace), Generative AI with LLMs (DeepLearning.AI), ML in Production (Coursera).