Looking for a job

Saurav Patel

@sauravpatel1

Message

ML researcher at IIIT Gwalior who surpassed an IEEE JBHI benchmark with a CNN–Transformer ensemble.

India

Message

At IIIT Gwalior, I extended the published CTransNet CNN–Transformer hybrid for medical image classification, reaching 98.78% accuracy on the BreaKHis dataset against the paper’s 98.29% benchmark. I also fused predictions from four magnification-specific models, reaching 100% binary accuracy and an AUC of 1.0000.

In my projects, I built ReviveAI, a multi-agent payment recovery system that achieved 3× the revenue recovery of blind retries in simulation. I’ve also developed an LSTM stock forecasting pipeline and StudyMate, a RAG-powered study assistant using LangChain and ChromaDB.

Experience

Work history, roles, and key accomplishments

Machine Learning Research

IIIT Gwalior

May 2026 - Jul 2026 (2 months)

Extended a published CNN-Transformer hybrid for medical image classification, surpassing the benchmark accuracy. Engineered a multi-magnification ensemble and resolved class imbalance to improve minority-class recall.

PyTorch Python NumPy scikit learn Git CNN Transformers Ensemble Methods Imbalanced Learning

Education

Degrees, certifications, and relevant coursework

Indian Institute of Information Technology and Management, Gwalior

Bachelor of Technology, Computer Science and Engineering

2023 -

Grade: 7.68/10.00

Pursuing a Bachelor of Technology in Computer Science and Engineering with a CGPA of 7.68/10.00, expected graduation in 2027.

Coursera

Machine Learning Specialization, Machine Learning

Completed the Machine Learning Specialization by Andrew Ng, covering supervised and unsupervised learning, neural networks, recommender systems, and reinforcement learning.