Bhavya Patel

AI Researcher

Jaipur, Rajasthan, India6 yrs 4 mos experience

AI EnabledHighly Stable

Key Highlights

Expert in building scalable AI pipelines.
Achieved 97% accuracy in intent detection.
Recognized with High Achiever Award for pipeline optimization.

Stackforce AI infers this person is a Machine Learning Engineer with a focus on Fintech and AI-driven solutions.

Contact

Skills

Core Skills

Machine LearningNatural Language Processing

Other Skills

AI AgentsAI pipelineAdobe LightroomAsuretifyBootstrapC (Programming Language)C++Cascading Style Sheets (CSS)Data StructuresDatabase Management System (DBMS)Document AIFlaskGenerative AIHTMLInstance segmentation

About

As a Machine Learning Engineer, I am passionate about leveraging AI that not only solve complex problems but also create meaningful value for end users. My current work involves building and optimizing pipelines for large language models (LLMs), deploying APIs into production and managing their performance to ensure they deliver value at scale. Beyond the technical aspects, I have a keen interest in strategic thinking, product analysis, and planning. I enjoy breaking down products to identify areas for improvement, understanding user needs, and combining the best features to create innovative solutions. My analytical mindset allows me to uncover insights that guide better decisions, whether in designing AI systems or enhancing product strategies. I thrive on exploring how technology and psychology intersect to influence business outcomes, and I’m driven by the opportunity to work on projects that blend innovation with purpose. #ContrarianEssence

Experience

Injala

2 roles

AI/ML Engineer

Aug 2023 – Present · 2 yrs 7 mos

💻 Working on Document AI
Collaborated on building a scalable, production-grade AI pipeline for end-to-end insurance document automation —combining page classification, field extraction, and tabular parsing, which significantly outperformed previous models in accuracy and reliability.
Built a robust intent detection module to interpret insurance coverage and endorsements, enabling contextual chatbot responses with 97%+ field-level accuracy.
Optimised a document intelligence pipeline to extract key data from specific PDF pages, cutting processing time from 50 seconds to 15–20 seconds, achieving over 65% improvement that directly enhanced system performance and user experience.
Developed a robust extraction framework capable of processing scanned, digitally generated PDFs, skewed, multi-column, and embedded-text PDFs, ensuring accurate and consistent data across varied document types.
Honored with the ‘High Achiever Award (Q2 2024) 🏆’ in recognition of high-impact contributions to pipeline optimization and user experience improvements.

Document AIAI pipelineintent detectiondocument intelligencedata extractionPDF processing+2