Available for Internships & Projects

Buildingintelligentproductsfromideatodeployment.

I'm AMJURI YASWANTH, a Full Stack & AI Developer specializing in production-grade RAG systems, multimodal AI, and full-stack web/mobile application development.

React NativeRAG SystemsNext.jsNode.jsMongoDBVision AI

View Projects

Contact Me

Résumé

0products shipped

0internships

0certifications

0%faster pages

Full Stack & AI

RAG & Vision Pipelines

AMJURI YASWANTH

Full Stack & AI Developer

Scroll

TECH_STACK // TOOLCHAIN

Core Development Stack

React

React Native

Next.js

Node.js

TypeScript

MongoDB

Express

RAG Systems

Vision AI

Tailwind CSS

Git & GitHub

Offline-First

AYtech_stack

01Developer Story

EngineeringtheintersectionofFullStack&IntelligentSystems.

My engineering philosophy centers around creating responsive, accessible user interfaces backed by intelligent algorithms. I focus on taking complex AI tools—such as vision classification, multimodal captioning, and custom RAG pipelines—and integrating them seamlessly into mobile and web systems.

Whether it is building multi-tenant SaaS backends with strict MongoDB isolation or designing offline-first architectures that execute local vector checks on the client, my focus is always on performance, scalability, and user-centric accessibility.

I thrive on end-to-end product development, from writing clean API routers to tuning page speed metrics by 30% to maximize user engagement.

Multimodal AI & Vision

Integrating deep learning models like EfficientNetB0 for issue categorization and BLIP for captioning directly into mobile client workflows.

Offline-First Design

Designing resilient pipelines and local DB caches that preserve critical app operations when cellular networks are unavailable.

Full Stack Scalability

Architecting multi-tenant databases, isolated API routers in Express, and stateful UIs utilizing React/React Native.

End-to-End Product Delivery

Shipping products from basic initial concept wireframes up to cloud deployments and production performance tuning.

02Competencies

TechnicalSkillsInventory

A curated inventory of the technologies I use to ship full-stack platforms and AI-powered products end to end.

0Categories

0Technologies

Frontend

90% proficiency

HTML
CSS
JavaScript
React
React Native
Bootstrap

Backend

85% proficiency

Node.js
Express.js
REST APIs

Database

82% proficiency

MySQL
MongoDB

Programming

88% proficiency

Java
SQL
JavaScript

AI / ML

86% proficiency

RAG Pipelines
Vector Search (TF-IDF)
Vision Model Integration
Roboflow

Tools

92% proficiency

Git
GitHub
VS Code
Linux
Expo

Concepts

90% proficiency

Responsive Design
Authentication
MVC Architecture
Offline-First Design

03Career Timeline

ProfessionalExperience

July 2025

AI & ML Intern

NisumConsulting

Worked on ML projects and collaborated with cross-functional teams to deliver business-aligned AI solutions.
Contributed to AI/ML development workflows and supported implementation of machine learning solutions.

May – June 2025

Full Stack Intern

TechnicalHub

Improved website page performance by 30% through frontend optimization and performance improvements.

Core Achievement

Innovation Award — Improved Website Speed by 30%

04Case Studies

Engineering,EndtoEnd

How I think and build — the problem, the approach, and the impact behind each project.

01Featured Case Study

NAGARA NETRA

AI-Powered Smart City Civic Grievance Platform

Report a civic issue with a photo, voice note, text, and GPS — one ticket, fully structured.

Product design, ML integration & full mobile stack

// problem

The problem

Civic complaint systems are text-only and frequently route issues to the wrong department. Citizens struggle to describe what they see, and agencies lose the location and visual context they need to act.

// approach

What I built

Built a React Native app that ingests four input modalities — photo, voice, text, and GPS — and runs a multimodal pipeline on-device: EfficientNetB0 classifies the issue, BLIP generates a caption, and MuRIL NLP normalizes the multilingual text into a structured grievance.

// impact

The result

Every report is auto-tagged and geo-localized, giving agencies a structured, searchable ticket with visual evidence instead of a free-text wall.

Multimodal Pipeline Architecture

Image Asset

Text Grievance

Voice Record

GPS Position

Classifier

EfficientNetB0 + BLIP

NLP Engine

MuRIL NLP

Output System

Structured civic ticket

Auto-tagged, geo-localized, stored in MongoDB

Fused image classification + captioning (EfficientNetB0 / BLIP) with MuRIL NLP for multilingual normalization
Four input modalities — image, text, voice, GPS — merged into a single structured ticket
On-device ML keeps processing fast and privacy-preserving

React NativeNode.jsMongoDBMuRIL NLPEfficientNetB0BLIP

Source

02Project Spotlight

VITANOVA AI

AI-Powered Nutrition Assistant

Scan a meal, get a multilingual nutrition breakdown — even when you're offline.

Solo builder — mobile app, RAG pipeline & offline architecture

// problem

The problem

Nutrition trackers rely on manual entry and die without connectivity — exactly when people eat at restaurants, cafeterias, and low-signal areas.

// approach

What I built

Scans a meal photo with Gemini vision, queries a local nutrition database through a custom RAG pipeline, and answers in multiple languages via voice interaction.

// impact

The result

Full nutrition breakdowns stay available offline, backed by a local vector cache that keeps answers fast, relevant, and private.

Offline-First RAG Architecture

Meal Photo

Voice Query

Nutrition Query

Offline Cache

Vision Model

Gemini Vision API

RAG Engine

Local Vector Cache

Output System

Macro Nutrition Specs

Offline-ready structured facts & voice output

Gemini Vision meal scanning converted into structured nutrition facts
Custom RAG pipeline over a local nutrition database
Offline-first design with multilingual voice interaction

React NativeExpoOpenRouterGemini VisionCustom RAG Pipeline

03Project Spotlight

MULTI-TENANT SAAS PLATFORM

SaaS backend engineered for strict tenant isolation and role-based control.

Architect & full-stack engineer

// problem

The problem

Multi-tenant SaaS collapses when one tenant can leak into another. Isolation had to be guaranteed at the database and API layers, not just the UI.

// approach

What I built

Engineered MongoDB multi-tenant isolation with tenant-scoped queries and modular role-based authorization, exposed through isolated Express routers per domain.

// impact

The result

A platform where tenants share infrastructure but never share data — with authorization enforced at every request boundary.

Tenant Isolation Architecture

Tenant Subdomain

Bearer Token

User RBAC Role

API Request

Routing Layer

Domain Routers

Authorization

Modular RBAC

Database Boundary

Tenant-Scoped Isolation

Query containment at MongoDB collection level

Tenant-scoped MongoDB isolation plus modular RBAC
Isolated Express routers for clean domain separation
React front-end with role-aware navigation and guards

ReactNode.jsMongoDB

Source

04Project Spotlight

ORYX

Skill Exchange Platform

A community where you share what you know and book what you don't.

Full-stack engineer

// problem

The problem

Skill-sharing communities need trust and discoverability; most fall apart on weak authentication or unusable navigation.

// approach

What I built

Built a React + Node.js + MongoDB platform with secure authentication, session-based skill exchange, and a fully responsive interface.

// impact

The result

A working skill marketplace where members find, book, and run sessions end-to-end.

Skill Exchange Platform Flow

Skill Sharers

Skill Seekers

Auth Token

Marketplace Query

Discovery

Matching Engine

Scheduler

Session Booking Core

Persistence Engine

Verified Skill Booking

Session state & exchange ledger in MongoDB

Secure authentication powering session-based skill exchange
Responsive, interactive front-end optimized for discoverability
REST API with MongoDB persistence

ReactNode.jsMongoDB

Source

05Academics

EducationBackground

2023 – 2027

B.Tech. in Computer Science and Engineering

Aditya College of Engineering and Technology

Result: CGPA: 8.56

2021 – 2023

Intermediate (12th)

Sasi Junior College

Result: 90%

Credentials

ProfessionalCertifications

Oracle Certified Foundations Associate (2025)

MongoDB Certified Associate Developer

GitHub Foundations

06Recognition

Achievements&Honors

Innovation Award — Improved Website Speed by 30%

Represented District & State Level in Badminton

Engagement

ExtracurricularActivities

NSS Coordinator

Lions Club Member

Community Volunteer

07Inquiries

GetInTouch

Open to internships & projects

Let's build something exceptional together.

I'm currently open to full-stack, mobile, and AI/ML internships and project collaborations. Tell me about your idea — I usually reply within 24–48 hours.

Email Address

yaswanthamjuri@gmail.com

Office Location

East Godavari, Andhra Pradesh, India

response within 24–48h

Résumé