Open to opportunities

Shiva Sai Vittal

@shivasaivittal

AI & ML engineering student building intelligent data classification, NLP systems, and automated data quality pipelines for enterprises.

India

Message

What I'm looking for

I’m looking for an early-career role where I can apply ML, NLP, and data quality automation to build enterprise-grade classification and analytics, with real ownership of AI pipelines and strong data governance impact.

I’m an AI & ML engineering student focused on turning messy, unstructured, and duplicate data into reliable, enterprise-ready information. I leverage machine learning, NLP, and data quality automation to deliver intelligent classification, master-data workflows, and AI-driven analytics.

In my projects, I built a Master Data Quality Automation pipeline that achieved 91% deduplication accuracy and reduced data errors by 78% through an automated cleansing and REST API integration. I also developed real-time abnormality detection with sub-second inference and a multi-model fusion approach that reduced false positives by 35%, plus an Intelligent Document Classifier using fine-tuned BERT that reached 94% F1-score and reduced manual routing time by 70%.

Experience

Work history, roles, and key accomplishments

Real-Time Abnormality Detection

Independent Projects

Engineered a real-time abnormal activity detection system on CCTV video streams using optimized computer vision models for violence, fire, and fall detection with sub-second inference. Reduced false positives by 35% via multi-model decision fusion (YOLOv8 + ResNet) and delivered production alerts through a Flask interface with secure SMTP notifications (99.9% delivery rate).

Python OpenCV TensorFlow YOLOv8 ResNet Computer Vision Flask

Intelligent Document Classifier

Independent Projects

Built an NLP-based document classification system to categorize unstructured enterprise documents (invoices, POs, contracts) into predefined taxonomy classes. Fine-tuned a BERT transformer model to achieve 94% F1-score across 12 categories, and implemented an OCR/tokenization preprocessing pipeline to reduce manual routing time by 70%.

Python BERT Transformers NLP OCR Text Preprocessing Tokenization Document Classification Data Pipelines

Master Data Quality Automation

Independent Projects

Built an automated data quality pipeline to detect and rectify duplicates, missing values, format inconsistencies, and outliers across structured enterprise datasets. Implemented NLP-based entity resolution with TF-IDF and cosine similarity (91% deduplication accuracy) and integrated a Flask REST API to reduce data errors by 78%.

Python Flask SQL BERT TF IDF Cosine Similarity Entity Resolution REST APIs

Education

Degrees, certifications, and relevant coursework

Chaitanya Bharathi Institute of Technology

Bachelor of Engineering, Artificial Intelligence and Machine Learning

2022 - 2026

Grade: CGPA 8.1 / 10

Pursuing a B.E. in Artificial Intelligence and Machine Learning with CGPA of 8.1/10. Coursework and projects include NLP-based document classification and master data quality automation using BERT and Flask.