001
> Software Developer_

ANIMESH SHARMA

Backend systems. Java ecosystem. Infrastructure that holds.

[scroll to initialize]
JAVASPRING BOOTPOSTGRESQLREST APILINUXSYSTEM DESIGNMICROSERVICESDOCKERCI/CDREACTPYTHONMONGODB JAVASPRING BOOTPOSTGRESQLREST APILINUXSYSTEM DESIGNMICROSERVICESDOCKERCI/CDREACTPYTHONMONGODB

          

> whoami

Animesh Sharma — Backend Engineer

Building reliable systems since the code compiled.

> cat /etc/profile

I write server-side software that processes, persists, and serves data at scale. My work lives in the JVM — Spring Boot services, PostgreSQL schemas, REST contracts. When something breaks at 2 AM, I want the system I built to tell me exactly why.

LOCATION
India
FOCUS
Backend Engineering
STACK
Java / Spring / PostgreSQL
ENVIRONMENT
Linux / Docker / Git
0PROJECTS SHIPPED
0TECHNOLOGIES
0LINES OF BACKEND CODE
01

BenefitBridge

Full-stack government welfare platform connecting 1000+ citizens to schemes. AI-powered eligibility matching, Three.js animations, JWT auth with role-based access, real-time analytics dashboard.

ReactDjangoPostgreSQLThree.jsJWT
[LIVE DEMO →]
02

Mavira OS

Custom Linux distribution built from scratch using Debian live-build. XFCE desktop with custom branding, AppArmor + Fail2Ban security hardening, bootable ISO with installer.

LinuxDebianBashAppArmorXFCE
[DOWNLOAD ISO →]
mavira-os — bash

$ _

03

Triage Assistant

AI-powered emergency triage assistant. Semantic protocol retrieval via ChromaDB + sentence-transformers, rule-based clinical reasoning. Returns triage level, next action, and risk scores via FastAPI backend with React scrollytelling frontend.

PythonFastAPIChromaDBReactAI
[LIVE DEMO →]
TRIAGE ASSESSMENT
CRITICAL9.2
URGENT7.8
SEMI-URGENT5.1
STANDARD2.4
> protocol_match: chest_pain_acute
> confidence: 0.94
> action: IMMEDIATE_ASSESSMENT
04

Health Coach AI

Production-grade agentic AI system for HPE GenAI hackathon. Multi-agent architecture, 94% data compression, budget-aware behavior modes, automatic fallback system, explainable AI decisions.

PythonAIMulti-AgentLLM
[VIEW CODE →]
USERI've been tracking 6 hours of sleep for the past week.
AGENTSleep debt accumulating. Multi-agent analysis: shifting wind-down 45min earlier. Budget mode: using cached recommendations. Confidence: 0.91
USERFactor in my morning workouts too.

Tooling

BACKEND
Java
Spring Boot
Hibernate
Maven
DATA
PostgreSQL
MongoDB
SYSTEMS
Linux
Docker
Bash
FRONTEND
JavaScript
React
HTML/CSS
TOOLS
Git
REST APIs
CI/CD

Let's build
something
precise.