AS.
Hello,

I'm Aditya Sharma. 

AI Engineer

Building intelligent software for real-world problems. Specialized in AI systems, RAG architectures, and production-ready full-stack applications.

GitHubLinkedInEmail
Scroll
01

About Me 

A brief look into my background and philosophy.

I am a final-year Computer Science student specializing in AI & Data Science at Poornima University. My passion lies at the intersection of complex machine learning systems and elegant product engineering.

I don't just build isolated models or experiments; I enjoy architecting complete, production-ready solutions. From designing scalable Retrieval-Augmented Generation (RAG) pipelines and predictive healthcare models to building the interactive Next.js and Streamlit interfaces that bring them to life.

Whether I'm integrating voice processing for an AI Call Bot or reducing data processing times by 60% with NLP interfaces, my goal is always to build software that solves practical business problems and delivers seamless user experiences.

15+

Projects Built

5+

AI Models Deployed

100+

Successful Merges

02

Capabilities 

Technologies and frameworks I use to build complete products.

AI Engineering & Data Science

Python
Machine Learning
Deep Learning
LLM Applications
RAG Systems
Prompt Engineering
Data Analysis
PyTorch
LangChain
Pinecone

Backend Development

FastAPI
REST APIs
SQL
MongoDB
PostgreSQL
Node.js

Frontend Engineering

Next.js
React
TypeScript
Tailwind CSS
Streamlit
Framer Motion
Zustand

Tools & Deployment

Docker
Git & GitHub
Vercel
Railway
Linux
GCP
03

Featured Projects 

A selection of my best work, integrating AI with full-stack development.

BLACKOUT

GitHub RepoLive Demo

Connectivity-Spectrum AI — Online, Offline, or Somewhere in Between

The ProblemAI assistants become useless without internet connectivity, leaving users stranded when they need help most — in remote areas, during outages, or on unstable networks.
The SolutionBuilt an adaptive AI system that seamlessly switches between Gemini 2.0 Flash (online) and a local Ollama model with RAG-augmented knowledge base (offline). Includes auto-detection, sync engine, and self-improving knowledge base through user feedback.
Next.js
FastAPI
Gemini API
Ollama
MongoDB
RAG
TypeScript
Python
Updated Jun 11, 2026

MindShield AI

GitHub RepoLive Demo

AI-powered mental wellness platform for students in competitive exams

The ProblemStudents preparing for high-stakes exams face burnout, anxiety, and mental health challenges with limited access to personalized support systems.
The SolutionDeveloped a comprehensive wellness platform featuring AI journal analysis, burnout radar, wellness analytics, an AI coach, and exam readiness scoring — all powered by Gemini AI with real-time data visualization.
Next.js 15
TypeScript
Supabase
Gemini API
Tailwind CSS
Recharts
Framer Motion
Updated Jun 6, 2026

CookPilot AI

GitHub Repo

AI meal planner with smart grocery lists and budget analysis

The ProblemDaily meal planning is time-consuming, often leads to food waste, and rarely accounts for budget constraints, dietary needs, and existing pantry items.
The SolutionCreated an AI-powered meal planning assistant with a 6-step wizard that generates personalized meal plans, smart grocery lists with cost estimates, ingredient substitutions, and cooking timelines — all optimized for budget and nutrition.
Next.js 15
TypeScript
Gemini 2.5 Flash
Zustand
Tailwind CSS
Zod
Updated Jun 6, 2026

Excel Chatbot

GitHub Repo

Natural language interface for spreadsheet data analysis and visualization

The ProblemNon-technical users struggle to extract insights from spreadsheet data, requiring knowledge of formulas, pivot tables, and chart creation.
The SolutionBuilt a conversational data analysis tool that accepts CSV/Excel uploads and answers natural language queries with auto-generated visualizations (bar, line, pie, scatter charts) and plain-English summaries using Gemini API.
Python
Streamlit
Gemini API
Pandas
Matplotlib
LangChain
Updated Apr 1, 2026

AI Call Bot

GitHub Repo

Automated voice assistant with sub-2-second response latency

The ProblemTraditional IVR systems frustrate callers with rigid menu trees, while human agents are expensive and inconsistent across high call volumes.
The SolutionImplemented a conversational voice agent using speech-to-text and LLM-based response generation, achieving sub-2-second response latency for real-time customer interactions via telephony integration.
Python
Speech Processing
NLP
LLMs
FastAPI
Updated Apr 6, 2025

AI Text Humanizer

GitHub Repo

Transform AI-generated text into natural, human-sounding content

The ProblemAI-generated content often reads mechanical and is easily flagged by detection tools, creating issues for content creators who use AI as a writing aid.
The SolutionBuilt a text transformation engine that rewrites AI-generated content to match natural human writing patterns while preserving meaning, tone, and context.
Python
NLP
LLMs
FastAPI
Updated Sep 9, 2025

GitHub Activity

Public Repos22
Followers3
Code FrequencyDaily

Top Languages

Python99%
TypeScript1%
04

Experience & Education 

My professional journey and academic background.

Apr 2026 — Present

AI & Machine Learning Intern

Jagatmitra Foundation & ICVAST Globe LLP • Delhi, India (Remote)
  • Architected a Medical RAG Chatbot with Pinecone vector databases, semantic search, and LLMs — reducing information retrieval time by 40%.
  • Designed predictive healthcare models (Cardiovascular Risk, Hypoglycemia) achieving ROC-AUC of 0.7206 for the Lipid-Driven Cardiac Risk Model.
  • Deployed scalable ML models via FastAPI and Streamlit, maintaining continuous integration across 15+ successful merges.
RAG
Pinecone
FastAPI
LLMs
Healthcare AI
Jan 2026 — Present

Freelance Data Annotator

Alignerr • Remote, USA
  • Validated over 5,000 data points for GenAI model training, improving dataset consistency and mitigating labeling errors in production workflows (RLHF).
  • Enhanced labeling accuracy by 25% through systematic quality audits.
RLHF
GenAI
Data Quality
Annotation
Jun 2025 — Jul 2025

Machine Learning Intern

Yeppar • Jaipur, Rajasthan
  • Developed NLP interfaces and analytical workflows using LangChain and LLMs, decreasing manual data processing time by 60%.
  • Built real-time Streamlit dashboards with semantic search, supporting 200+ daily queries.
LangChain
NLP
Streamlit
Data Visualization
2023 — 2027

B.Tech in Computer Science Engineering (AI & Data Science)

Poornima University • Jaipur, Rajasthan
  • Final-year student specializing in AI & Data Science with a CGPA of 7.85.
  • Focus areas: Machine Learning, Deep Learning, NLP, Computer Vision, and RAG systems.
AI & Data Science
Computer Science
CGPA: 7.85
05

Certifications 

Professional credentials and cloud achievements.

Oracle Cloud Infrastructure AI Foundations Associate

Oracle

2024Verify

GCP Cloud Practitioner

Google Cloud

2024Verify

MongoDB Developer Fundamentals

MongoDB

2024Verify
06

Get In Touch 

Open to exciting opportunities, freelance projects, and collaborations.

Let's build something great.

I'm currently looking for internships and full-time roles in AI Engineering and Full-Stack Development. Whether you have a question or just want to say hi, I'll try my best to get back to you!

Location

Jaipur / Delhi, India