LOADING...

000%

Full Stack · ML Engineer · VIT Chennai

Shivam Chavan

Shivam Chavan

Building|
Docker
PostgreSQL
Python
PyTorch
FastAPI
Next.js
React
Scroll

Crafting the Future
One Line at a Time

Hey! I'm Shivam Sandip Chavan - a CS engineer who loves building things that actually matter. From AI-powered document systems to production-grade cloud apps, I bridge the gap between research and shipping.

I've interned at startups building real products, published research in deep learning for medical imaging, and filed a patent in autonomous vehicle navigation - all while maintaining a top-tier CGPA at VIT Chennai.

0.00
CGPA
0+
Patents & Publications
0+
Years Experience
0+
Major Projects

Experience.

A timeline of my professional journey, building production systems and shipping high-impact products.

Full-time

Full Stack Developer

PenPixel LLP
Jun 2025 – Sep 2025
  • Architected and developed a comprehensive Employee Onboarding System using Next.js and MongoDB, significantly streamlining internal HR workflows.
  • Automated complex document processing and form-filling workflows using Puppeteer, reducing manual onboarding effort and processing time.
  • Deployed scalable production infrastructure on Hetzner Cloud using Coolify, while standardizing backend services with Swagger API docs.
Next.jsMongoDBPuppeteerCoolifyHetznerSwagger
Internship

Product Intern

Ziga Infotech
Mar 2025 – Jun 2025
  • Engineered a high-performance Flutter mobile application tailored for real-time industrial image processing and automated object detection.
  • Integrated the advanced YOLOv8 computer vision model to detect and map pipes in real-time, achieving an operational accuracy of 95%.
  • Collaborated with a cross-functional team of 3 to seamlessly bridge frontend systems with backend processing, optimizing model deployment.
FlutterYOLOv8PythonComputer VisionREST APIs
Freelance Project

Freelance Frontend Developer

AAS Tour & TravelsVISIT
Jun 2026
  • Designed a premium, luxury-themed single-page website featuring a cinematic dark aesthetic and a custom animated splash screen.
  • Implemented complex scroll-based animations using AOS while ensuring robust client-side form validation for user interactions.
  • Built a clean, responsive component architecture focused on maximum accessibility and high-performance cross-device rendering.
Next.js 16React 18TypeScriptTailwind CSS v4AOS
Source Code
Freelance Project

Freelance Web Developer

Sakaar Homoeopathic ClinicVISIT
Dec 2024 – Jan 2025
  • Developed and deployed a modern digital platform tailored to improve patient engagement and showcase specialized clinical treatments.
  • Engineered a secure online appointment booking system featuring automated real-time email notifications integrated via EmailJS.
  • Managed the end-to-end project lifecycle, from initial client requirement gathering to final production deployment on Firebase.
HTML5CSS3JavaScriptTailwind CSSFirebaseEmailJS
Source Code

Projects.

Real-world AI and full-stack systems with measurable impact.

1 / 4
Jul 2025

LLM-Powered Intelligent Document Query System

An AI-powered document query system using RAG and intelligent agents for insurance, legal, HR, and compliance use cases.

  • Processes 15+ file formats with OCR and secure validation for robust document handling.
  • Integrated Sentence Transformers + FAISS for semantic search, achieving 92% accuracy.
  • Built a scalable pipeline using FastAPI, Mistral API, and HuggingFace, ensuring reliable responses in <10s latency.
RAGFastAPIMistral APIHuggingFaceFAISSPythonOCR
1 / 4

← → arrow keys to navigate

Publications

Research & Innovation

Contributing to the frontier of AI and Computer Vision research

📋 PatentFeb 2025

System And Method For Autonomous Navigation Of Vehicle

Patent ID: 202541013968

Constructed an autonomous self-driving car using computer vision, leveraging lane detection, YOLO-based object recognition, and low-latency processing pipelines for real-time obstacle avoidance and decision-making.

Lane DetectionYOLO Object RecognitionLow-latency Processing
📄 Conference PaperSep 2025

Parkinson Image Synthesis and Classification using Deep Learning

ICEC2NT 2025

Developed a GAN-based deep learning framework for Parkinson's Disease detection, generating synthetic medical images to address data scarcity and improve classification performance.

98% AccuracyGAN ArchitectureMedical Imaging
Skills

My Tech Arsenal

Tools and technologies I bring to every project

{ }

Languages

6
JavaCC++PythonJavaScriptSQL

Frameworks

5
ReactJSNext.jsFlutterFlaskTailwind CSS
🧠

AI & ML Libraries

5
TensorFlowOpenCVKerasHuggingFaceFAISS
☁️

Databases & Cloud

4
MySQLMongoDBFirebaseSupabase
🔧

Developer Tools

7
GitHubREST APIsSwaggerPuppeteerCoolifyHetznerKaggle
Education

Academic Foundation

VIT

Vellore Institute of Technology, Chennai

B.Tech in Computer Science and Engineering
Aug 2023 – Present
0.00
CGPA / 10
Student Council Member (2025–26)CGPA: 9.28 / 10
Contact

Let's Build Something
Amazing Together

I'm open to full-time roles, internships, freelance projects, and research collaborations. Don't hesitate to reach out!