Founder @ Rewrite Labs · Final Year CS Student · Shipping real AI products

github

M. Adhitya  • Peerlist

<h3>Unlock Deeper Insights with DataPulse</h3>
<p>DataPulse is a powerful data analytics platform designed to transform raw data into actionable insights. It empowers businesses to understand trends, identify opportunities, and make data-driven decisions with confidence.</p>
<p><strong>Key Features:</strong></p>
<ul>
<li>Intuitive dashboards for visualizing complex data.</li>
<li>Advanced analytical tools for in-depth exploration.</li>
<li>Real-time data processing and reporting.</li>
<li>Customizable reports tailored to your specific needs.</li>
<li>Seamless integration with various data sources.</li>
</ul>
<p>Leverage the full potential of your data with DataPulse, your comprehensive solution for modern data analytics. Whether you're looking to optimize marketing campaigns, understand customer behavior, or improve operational efficiency, DataPulse provides the clarity and tools you need to succeed.</p>

DataPulse — Data Analytics

Co-developed a Random Forest-based Machine Learning model to automate fraud detection. Collaborated on code structure and algorithm optimization within a Python environment to ensure robust data analysis.

Credit Card Fraud Detection Model

📊 Customer Segmentation using K-Means Clustering — AI Internship @ SmartED Innovations One of the most practical ML problems in retail: who are your customers, really? For my minor project at SmartED Innovations, I built a complete customer segmentation pipeline using K-Means clustering — identifying 5 distinct buyer personas from a 200-customer retail dataset. What the model found: → 🟢 Cluster 3 — High income + high spenders → VIP targets for loyalty programmes → 🔵 Cluster 1 — High income + moderate spend → Upsell candidates → 🔴 Cluster 2 — Low income + high spend → Impulsive buyers, respond to flash sales → 🟠 Cluster 0 — High income + low spend → Conservative savers, need trust-building → 🟣 Cluster 4 — Low income + low spend → Budget shoppers, discount-sensitive Technical pipeline: → Feature selection: Annual Income + Spending Score → StandardScaler for feature normalisation → Elbow Method to determine optimal k=5 → K-Means++ initialisation for stable convergence → Seaborn scatter plot with centroid markers for visual validation Tech: Python · pandas · scikit-learn · matplotlib · seaborn 🔗 GitHub: github.com/iamadhitya1/customer-segmentation-retail #MachineLearning #Python #KMeans #CustomerSegmentation #DataScience #UnsupervisedLearning #RetailAnalytics #AIInternship #SmartEDInnovations #IITRAM

Customer Segmentation Retail

FocusVision is a real-time computer vision productivity tool that uses your webcam to detect and alert you the moment a distraction enters your environment — all without a single line of server code. It runs an object detection model (COCO-SSD) directly in the browser using TensorFlow.js, scanning your camera feed continuously and firing audio alerts and browser push notifications when it spots a phone, food, a second person in frame, a TV, or notices you've left your seat. Every session is tracked with a live focus score (% of time spent undistracted), a distraction counter with per-category breakdown, a longest uninterrupted streak timer, and a configurable goal (e.g. 25-minute Pomodoro). On session end, a full summary is generated and saved to session history (up to 30 sessions stored locally). A Ghost Mode hides the camera feed while keeping detection active — for users who find the live view distracting itself. Each distraction type can be toggled on or off independently, and alerts can be snoozed for 5 minutes. No API keys, no backend, no data ever leaves the device. The COCO-SSD model is loaded once and runs on the GPU via WebGL through TensorFlow.js. Tech: React · TensorFlow.js · COCO-SSD (Object Detection) · Web Audio API · Notifications API · WebRTC (getUserMedia) · Tailwind CSS · Vite

FocusVision — AI Distraction Detection for De

🧠 Built an A Pathfinding System from Scratch — AI Internship @ SmartED Innovations* During my AI Internship at SmartED Innovations (Dec 2025), I implemented the A* Search Algorithm in Python — one of the most fundamental and widely used algorithms in AI, robotics, and game development. What it does: → Navigates a 10×10 grid from Start to Goal while avoiding obstacles → Uses Manhattan distance as the heuristic — admissible, consistent, and optimal → Built with a min-heap priority queue (Python's heapq) for efficient node expansion → Fully visualized with Matplotlib — green path markers, obstacle walls, start & goal What I learned: → How A* balances exploration efficiency with path optimality using f(n) = g(n) + h(n) → Why heuristic choice (Manhattan vs. Euclidean) matters for correctness on grids → Clean OOP design — separating the environment, algorithm, and visualization → The difference between completeness and optimality in graph search Tech: Python 3 · heapq · Matplotlib · OOP 🔗 GitHub: github.com/iamadhitya1/intelligent-pathfinding-astar #ArtificialIntelligence #Python #Pathfinding #Algorithms #MachineLearning #AIInternship #SmartEDInnovations #IITRAM #ComputerEngineering

Intelligent Pathfinding A Star

Lectura.AI — AI-Powered Study Companion Lectura.AI is a browser-based academic assistant that transforms how students capture and retain knowledge. It features real-time lecture transcription using the Web Speech API, instant AI-formatted study notes, PDF/DOCX document analysis with Q&A chat, and auto-generated flashcards and quizzes — all running client-side with no backend or data storage. Built with React, Vite, and Tailwind CSS, the app integrates with Groq (LLaMA 3.3 70B) and Google Gemini via user-supplied API keys, ensuring complete data privacy. Documents are parsed entirely in-browser using PDF.js and Mammoth.js — only extracted text ever reaches the AI. Tech: React · Vite · Tailwind CSS · Groq API · Google Gemini API · Web Speech API · PDF.js · Mammoth.js

Lectura.AI - The AI Powered Study Assistant

**Library Management System Project | Java & OOP Concepts** During my 3rd semester at IITRAM, Ahmedabad, I developed a library management system using Java, incorporating core object-oriented programming principles like inheritance and abstraction. This project strengthened my programming fundamentals and problem-solving skills, as I designed features to efficiently handle book records, user data, and borrowing history. It was a hands-on experience that deepened my expertise in software development and code architecture—skills I continue to build on today. GitHub - https://github.com/iamadhitya1/library-management-system

Library Management System Using Java

🧪 Rewrite Labs — Built This from Scratch at 20. Here's What It Is. Rewrite Labs is a personal initiative I started to solve a problem I kept seeing around me: engineers who are technically excellent but personally stuck — running someone else's script, not their own. So I built a platform around one idea: your career is code, and it can be rewritten. What Rewrite Labs is: It's not just a website. It's a three-part ecosystem I built entirely on my own initiative — → 📰 Code & Context — a LinkedIn newsletter I run with 3,200+ engineer subscribers, covering career systems, mindset, and tech → 📖 The Rewrite — a 70-page book I authored (12 case studies of people who rewrote their own lives) → 💻 The Rewrite OS — a SaaS tool I designed and shipped (React 19 + Firebase + Framer Motion), live at therewriteos.netlify.app → 🌐 The Hub — the landing site at rewrite-labs.netlify.app tying it all together What I built the site with: React 19, Vite, React Router DOM v7, Framer Motion, Firebase, Netlify — dark terminal aesthetic because engineers deserve a landing page that actually feels like their world. The reach so far: 3,000+ engineers reached · Featured by BeerBiceps SkillHouse (one of India's largest creator-education platforms) · 3,200+ newsletter subscribers · 70-page book published This is not a college project. No mentor assigned it. No deadline pushed it. I identified a gap, designed the system, and shipped it. That's what Rewrite Labs is — my proof that the best projects are the ones you build before anyone tells you to. 🔗 https://rewritelabs.vercel.app/ #RewriteLabs #PersonalProject #BuildInPublic #ReactJS #Firebase #SaaS #EngineerMindset #BeerBiceps #Newsletter #IITRAM #SelfInitiated #Founder

Rewrite Labs

The Problem: The current engineering education model acts like a "Legacy Operating System." It optimizes for vanity metrics (grades/attendance) rather than production skills, leading to "Tutorial Hell" and early career burnout. The Solution: Rewrite Labs is an educational research initiative designed to debug the student mindset. The Build: Architecture: Designed and deployed a high-performance, responsive web platform using React.js (Vite) and Tailwind CSS. Design System: Implemented a custom "Terminal/Cyberpunk" aesthetic to align with the "System Rewrite" brand identity. Content: Hosts "The Rewrite" (v1.0), a 70-page manual analyzing behavioral patterns in engineering careers, currently being read by 2,000+ students. Status: Live & Deployed (v1.0)

Rewrite Labs | The Engineering Career Operati

I followed the step-by-step process of the SDLC life cycle of software development with my team members and illustrated how we will develop our service for Venue booking and event management adopting Behavior Driven Approach

Smart Event Planner & Management System

📖 I Wrote a Book. Here's Why. "The Rewrite: 12 Lives That Escaped the Script" — authored, published, and self-released in 2025. Not a textbook. Not a journal. A book built on one question: what if the life you're living was never actually yours? Most people run on default settings — the script handed to them by parents, schools, and society. This book is about people who audited that script, found the bugs, and deployed a new version of themselves. 12 chapters. 12 fictional people. 12 real problems you've probably felt: → Aryan — the perfectionist who confused high standards with fear → Nia — the influencer who realized applause wasn't the same as worth → Vikram — the coder who had the ideas but wouldn't execute → Saanvi — who learned that boundaries aren't rude, they're rescue lines → Major Dev — a soldier who fought everyone but himself → Anika — a CEO who lost everything and found that ego was the problem → Kai — a tech genius who finally cracked his own inner code → ...and 5 more Each story reads like a case study. Themes: atomic habits, ego death, self-advocacy, vulnerability, digital detox, purpose, burnout, and the courage to choose your own life. The framing is deliberately technical — debugging, source code, operating system — because I wrote this for people like me: engineers and builders who think in systems but forget to apply that logic to themselves. The Rewrite is available now. 🔗 Read more: linkedin.com/in/loveadhitya/ #TheRewrite #SelfPublished #PersonalDevelopment #Author #WrittenByAnEngineer #IITRAM #MindsetShift #BookLaunch #IndieAuthor #BuilderMindset

The Rewrite - Book by M Adhitya

Unclause is a browser-based legal document analyzer that makes contracts and agreements understandable to anyone — no law degree required. Users upload any PDF or DOCX (employment contracts, terms of service, NDAs, rental agreements), and the app instantly generates a plain English summary, flags risky or one-sided clauses with severity ratings (High / Medium / Low), and opens a Q&A chat so users can ask specific questions like "Can I cancel anytime?" or "What are the penalties in clause 4?" The entire pipeline runs client-side — documents are parsed in-browser using PDF.js and Mammoth.js, with only the extracted text reaching the AI. There is no backend, no database, and no data retention. Users supply their own free Groq or Gemini API key, which lives in browser memory and is cleared the moment the tab closes. Tech: React 19 · Vite · Tailwind CSS v4 · Groq API (LLaMA 3.3 70B) · Google Gemini API · PDF.js · Mammoth.js