Applied AI and Computer Vision Engineer

Vansh Patel

Building production-grade AI systems across computer vision, deep learning, MLOps, and deployment-ready intelligent monitoring.

Live Focus

Applied AI engineer building real-world intelligent systems.

View Projects Resume GitHub LinkedIn

Base Layer

Ahmedabad, Gujarat, India

Core Focus

Applied AI

Primary Domain

Computer Vision

Deployment Mindset

Production-grade

Location

Ahmedabad, India

System Signature

Future-facing AI systems with research depth and production intent.

About

An applied AI profile shaped like a modern product system.

The new direction is less flashy-template and more premium product design: quieter surfaces, stronger hierarchy, cleaner contrast, and sharper storytelling around systems work.

Third-year Computer Science undergraduate specializing in machine learning, deep learning, computer vision, and AI systems engineering.

Builds practical AI systems spanning real-time inference, tracking systems, transformer architectures, deployment pipelines, and intelligent monitoring.

Focused on shipping scalable, real-world intelligent systems instead of isolated research prototypes.

Tech Stack

A stack tuned for practical inference, deployment, and model operations.

Grouped like a modern systems portfolio rather than a checklist, with just enough visual weight to feel premium without turning noisy.

Programming

Core tooling cluster

PythonC++JavaSQL

Machine Learning

Core tooling cluster

PyTorchTensorFlowScikit-LearnXGBoost

Computer Vision

Core tooling cluster

YOLO11OpenCVVision TransformersByteTrack

Backend and Deployment

Core tooling cluster

FastAPIDockerFirebaseAWS

Tools

Core tooling cluster

GitLinuxVS Code

Featured Projects

A project-first portfolio built around system depth.

This redesign leans into what is actually in demand right now: strong editorial hierarchy, restrained luxury color, real screenshots, and cleaner proof of engineering capability.

LeafCure - Multi-Crop Disease Detection System

Real-time inference

Research-grade architecture

Platform home

Prediction results

Research mode

Flagship research system

LeafCure - Multi-Crop Disease Detection System

A production-grade AI crop disease detection platform built with Vision Transformers for real-time classification across 38+ crop disease classes.

PyTorchVision TransformersFastAPIReactFirebase

Features

Vision Transformer based classification
Real-time disease prediction
FastAPI inference services
Full-stack deployment architecture
Analytics dashboard
User management
Cloud-ready scalable deployment

Highlights

End-to-end AI system
Real-world deployment architecture
Transformer-based computer vision
Research-oriented implementation
Submitted research paper

Research Note

Research paper based on hierarchical attention mechanisms and deep learning systems submitted to IEEE Access.

GitHub Resume

System Canvas

ForgeGuard - Industrial Safety Intelligence System

Detect

Track

Respond

Industrial monitoring

ForgeGuard - Industrial Safety Intelligence System

A real-time industrial safety intelligence system for construction environments integrating object detection, tracking, hazard-zone analysis, and temporal monitoring.

YOLO11ByteTrackOpenCVPyTorch

PPE detection

Multi-object tracking

Hazard-zone monitoring

Compliance intelligence

Temporal violation detection

GitHub

System Canvas

MindScan - AI Mental Health Detection System

Detect

Track

Respond

Healthcare AI

MindScan - AI Mental Health Detection System

An AI healthcare monitoring platform using CNN-LSTM architectures for ECG-based mental health analysis and real-time physiological monitoring.

CNN-LSTMFlaskWebSockets

ECG signal classification

Real-time monitoring

WebSocket infrastructure

Edge-device optimization

Raspberry Pi deployment

GitHub

Experience

Machine learning work with measurable business lift.

The experience section is kept tight and credible, which tends to read much more premium than overstated timelines.

Codveda Technologies

Machine Learning Engineer Intern

Built churn prediction pipeline using XGBoost

Improved accuracy from 80% to 95%

Applied SMOTE and feature engineering

Built clustering systems using PCA and K-Means

Developed production-ready ML workflows

Achievements

Competitive signals, research momentum, and credibility markers.

Designed as strong supporting proof, without overpowering the project-led narrative.

Top 5% on Kaggle

Placed in the top 150 among 3000+ teams through competitive ML problem solving.

National Finalist - Infotsav '24

Recognized nationally for building a high-impact technical solution under competition pressure.

IEEE Access Submission

Submitted research work focused on hierarchical attention mechanisms and deep learning systems.

Industry Certifications

Earned certifications from IBM, DeepLearning.AI, and Deloitte across data and AI domains.

GitHub

A live engineering layer, styled to match the rest of the portfolio.

The visuals now follow the same graphite and metallic direction instead of clashing with bright blue stats cards.

Contribution Graph

GitHub Stats

Top Languages

Contact

Let's build intelligent systems with real-world impact.

Available for AI engineering collaborations, research-oriented builds, and system design work centered on computer vision and production ML.

GitHub

Location

Download Resume patelvanshj.17@gmail.com